Front Cover
 Title Page
 Steering committee
 Table of Contents
 Cultural and socioeconomic...
 Postharvest food loss assessment...
 Cereal grains and grain legume...
 Postharvest losses of fish
 Education, training, and exten...
 Conclusions and recommendation...
 Organizations actively involved...
 Board on Science and Technology...
 Back Cover

Title: Postharvest food losses in developing countries
Full Citation
Permanent Link: http://ufdc.ufl.edu/UF00087168/00001
 Material Information
Title: Postharvest food losses in developing countries
Series Title: Postharvest food losses in developing countries
Physical Description: viii, 206 p. : ill. ; 23 cm.
Language: English
Creator: National Research Council (U.S.) -- Board on Science and Technology for International Development
Publisher: National Academy of Sciences
Place of Publication: Washington
Publication Date: 1978
Subject: Food supply -- Developing countries   ( lcsh )
Food crops -- Postharvest losses -- Developing countries   ( lcsh )
Food -- Storage   ( lcsh )
Crops, Agricultural   ( mesh )
Developing Countries   ( mesh )
Food Contamination   ( mesh )
Food Preservation   ( mesh )
Aliments -- Entreposage   ( rvm )
Aliments -- Approvisionnement -- Pays en voie de développement   ( rvm )
Cultures vivrières -- Pays en voie de développement   ( rvm )
Alimentos, Abastos de -- Países subdesarrollados
Genre: bibliography   ( marcgt )
non-fiction   ( marcgt )
Statement of Responsibility: Board on Science and Technology for International Development, Commission on International Relations, National Research Council.
Bibliography: Includes bibliographies.
General Note: Text in English with summaries in French and Spanish.
 Record Information
Bibliographic ID: UF00087168
Volume ID: VID00001
Source Institution: University of Florida
Rights Management: All rights reserved by the source institution and holding location.
Resource Identifier: oclc - 04564626
lccn - 78070607

Table of Contents
    Front Cover
        Front Cover 1
        Front Cover 2
        Errata 1
        Errata 2
    Title Page
        Page i
        Page ii
    Steering committee
        Page iii
        Page iv
        Page v
        Page vi
    Table of Contents
        Page vii
        Page viii
        Page 1
        Page 2
        Page 3
        Page 4
        Page 5
        Page 6
        Page 7
        Page 8
        Page 9
        Page 10
        Page 11
        Page 12
        Page 13
        Page 14
    Cultural and socioeconomic aspects
        Page 15
        Page 16
        Page 17
        Page 18
        Page 19
        Page 20
        Page 21
        Page 22
        Page 23
        Page 24
        Page 25
    Postharvest food loss assessment and estimation
        Page 26
        Page 27
        Page 28
        Page 29
        Page 30
        Page 31
        Page 32
        Page 33
        Page 34
        Page 35
        Page 36
        Page 37
        Page 38
        Page 39
        Page 40
        Page 41
        Page 42
        Page 43
        Page 44
        Page 45
        Page 46
    Cereal grains and grain legumes
        Page 47
        Page 48
        Page 49
        Page 50
        Page 51
        Page 52
        Page 53
        Page 54
        Page 55
        Page 56
        Page 57
        Page 58
        Page 59
        Page 60
        Page 61
        Page 62
        Page 63
        Page 64
        Page 65
        Page 66
        Page 67
        Page 68
        Page 69
        Page 70
        Page 71
        Page 72
        Page 73
        Page 74
        Page 75
        Page 76
        Page 77
        Page 78
        Page 79
        Page 80
        Page 81
        Page 82
        Page 83
        Page 84
        Page 85
        Page 86
        Page 87
        Page 88
        Page 89
        Page 90
        Page 91
        Page 92
        Page 93
        Page 94
        Page 95
        Page 96
        Page 97
        Page 98
        Page 99
        Page 100
        Page 101
        Page 102
        Page 103
        Page 104
        Page 105
        Page 106
        Page 107
        Page 108
        Page 109
        Page 110
        Page 111
        Page 112
        Page 113
        Page 114
        Page 115
        Page 116
        Page 117
        Page 118
        Page 119
        Page 120
        Page 121
        Page 122
        Page 123
        Page 124
        Page 125
        Page 126
        Page 127
        Page 128
        Page 129
        Page 130
        Page 131
        Page 132
        Page 133
        Page 134
        Page 135
        Page 136
        Page 137
        Page 138
        Page 139
    Postharvest losses of fish
        Page 140
        Page 141
        Page 142
        Page 143
        Page 144
        Page 145
        Page 146
        Page 147
        Page 148
        Page 149
        Page 150
        Page 151
        Page 152
        Page 153
        Page 154
        Page 155
        Page 156
        Page 157
        Page 158
    Education, training, and extension
        Page 159
        Page 160
        Page 161
        Page 162
        Page 163
        Page 164
        Page 165
    Conclusions and recommendations
        Page 166
        Page 167
        Page 168
        Page 169
        Page 170
        Page 171
        Page 172
        Page 173
        Page 174
        Page 175
        Page 176
        Page 177
        Page 178
        Page 179
        Page 180
        Page 181
    Organizations actively involved in postharvest food conservation
        Page 182
        Page 183
        Page 184
        Page 185
        Page 186
        Page 187
        Page 188
        Page 189
        Page 190
        Page 191
        Page 192
        Page 193
        Page 194
        Page 195
        Page 196
        Page 197
        Page 198
        Page 199
        Page 200
        Page 201
        Page 202
        Page 203
        Page 204
    Board on Science and Technology for International Development (BOSTID)
        Page 205
        Page 206
        Page 207
    Back Cover
        Page 208
Full Text



Siii '1 i--i-"
11 1 He
11. . _R H




In Table 4.2, page 65, the reported losses of rice
in India should read:

6.0 percent
3 5.5 percent

1.0 percent

Unspecified Storage
Traditional On-Farm Storage
(Boxall and Greeley, 1978)
Improved On-Farm Storage
(Boxall and Greeley, 1978)

Postharvest Food Losses in
Developing Countries


Food Losses in



Board on Science and Technology
for International Development
Commission on International Relations
National Research Council

Washington, D.C. 1978

This report has been prepared by an ad hoc advisory panel of the Board on Science and
Technology for International Development, Commission on International Relations,
National Research Council, for the Office of Agriculture, Bureau for Development Sup-
port, Agency for International Development, Washington, D.C., under Contract No.
AID/csd-2584, Task Order No. 23.

NOTICE: The project that is the subject of this report was approved by the Governing
Board of the National Research Council, whose members are drawn from the Councils of
the National Academy of Sciences, the National Academy of Engineering, and the Insti-
tute of Medicine. The members of the Committee responsible for the report were chosen
for their special competence and with regard for appropriate balance.
This report has been reviewed by a group other than the authors according to pro-
cedures approved by a Report Review Committee consisting of members of the National
Academy of Sciences, the National Academy of Engineering, and the Institute of

Library of Congress Catalog Number 78-70607

Steering Committee for Study on
Postharvest Food Losses in Developing Countries

E. R. PARISER, Department of Nutrition and Food Science, Massachusetts
Institute of Technology, Cambridge, Massachusetts 02139, USA, Chairman
EDWARD S. AYENSU, Office of Biological Conservation, Smithsonian Insti-
tution, Washington, D. C. 20560, USA
MALCOLM C. BOURNE, New York State Agricultural Experiment Station,
Comell University, Geneva, New York 14456, USA
DWIGHT S. BROTHERS, Fairhaven Hill, Concord, Massachusetts 01742,
A. A. C. HUYSMANS, Action Programme for the Prevention of Food Losses,
Agriculture Department, Food and Agriculture Organization of the United
Nations, Rome, Italy
JOHN PEDERSEN, Department of Grain Science and Industry, Kansas State
University, Manhattan, Kansas 66502, USA
WILLIAM L. RATHJE, Department of Anthropology, University of Arizona,
Tucson, Arizona 85721, USA
P. C. SPENSLEY, Tropical Products Institute, London, England
GEORGE F. SPRAGUE, Department of Agronomy, University of Illinois,
Urbana, Illinois 61801, USA
DAISY M. TAGLIACOZZO, Department of Sociology, University of Massa-
chusetts, Harbor Campus, Boston, Massachusetts 02125, USA
M. G. C. McDONALD DOW, Board on Science and Technology for Interna-
tional Development, National Academy of Sciences, Washington, D.C.
20418, USA, Study Staff
JOHN G. HURLEY, Board on Science and Technology for International
Development, National Academy of Sciences, Washington, D.C. 20418,
USA, Study Staff


The Committee on Postharvest Food Losses in Developing Countries re-
ceived help from many sources and wishes to thank all who contributed to
the report. A list of contributors in many parts of the world who sent infor-
mation about postharvest food losses and who participated in working meet-
ings is provided at the end of the report.
The Committee and Staff also express their particular appreciation to the
following people who contributed to the planning of the study and the prep-
aration of the report and its review, either themselves or with the assistance
of their colleagues: D. G. Coursey, P. F. Prevett, and P. Tyler of the Tropical
Products Institute, London; G. G. Corbett, D. G. James, and H. Shuyler of
the Food and Agriculture Organization of the United Nations; K. Harris and
C. Lindblad, American Association of Cereal Chemists, Bethesda, Maryland;
N. L. Brown, Department of Energy, Washington, D.C.; E. Colson, University
of California, Berkeley; R. Davis, U.S. Department of Agriculture, Savannah,
Georgia; R. Fanale, American Association for the Advancement of Science,
Washington, D.C.; R. Forrest, University of Alberta; W. Furtick, University of
Hawaii; G. Harrison, University of Arizona; S. Hendricks, Silver Spring,
Maryland; E. Knipling, U.S. Department of Agriculture, Beltsville, Maryland;
M. Milner, Massachusetts Institute of Technology; R. Phillips, Kansas State
University; D. Pimentel, Comell University; D. Pickering, International Bank
for Reconstruction and Development, Washington, D.C.; R. Revelle, Univer-
sity of California, San Diego; A. Roetzer, St. Regis Paper Company, West
Nyack, New York; M. Whiting, Washington, D.C.; and to K. Byergo and W.
Smith Greig of the Agency for International Development, Washington, D.C.,
for their cooperation.
Our thanks also to the study staff: to F. R. Ruskin, who edited and pre-
pared the manuscript for publication, for the immense improvement in clarity
and brevity that resulted from her efforts with a difficult and diverse subject;
to R. Morris and M. Risdon who prepared the bibliography and contributed in
other ways; to T. Bass for administrative support; and to D. Mojka, Y. Cassells,
and A. T. Chau who cheerfully typed many drafts.


Comments on this report, especially regarding its usefulness and any initia-
tives or research it may have induced, would be welcomed by the staff of-
ficers: M. G. C. McDonald Dow and John Hurley, National Academy of Sci-
ences-National Research Council, 2101 Constitution Avenue N.W., JH-214,
Washington, D.C. 20418, USA.


Loss Estimation
Loss Reduction

Charge and Objective
Scope of the Study
Definitions and Boundaries




Cultural and Social Factors in Food Conservation
Economic Factors in Food Conservation
A Special Approach to Cost-Benefit Analysis

Relationship Between Accuracy and Usefulness of
Loss Assessment
Loss Estimation Methodology
Recommended Reading

General Causes of Postharvest Grain Loss
Individual Crop Loss Problems
Grain Legumes

Recommended Reading





Estimates of Loss
Causes of Loss
Preservation, Storage, and Conservation
Perishable Crops
Recommended Reading

General Aspects of Fish Harvest and Consumption
Postharvest Losses
Strategies for Loss Reduction
Recommended Reading

Education and Training
Extension Services
Recommended Reading

General Conclusions



Rdsume en Francais

Resumen en Espafiol

BOSTID Publications

Board on Science and Technology for International
Development (BOSTID)


As the world's population grows, increasing the food supply becomes an
ever-more-urgent priority. One vital-and neglected-step toward this end is to
reduce the food losses that occur between harvest and consumption. Reliable
studies indicate that postharvest losses of major food commodities in develop-
ing countries are enormous, in the range, conservatively, of tens of millions of
tonnes* per year and valued at billions of dollars. Programs for reducing these
losses must be based on reasonable estimates of their magnitude, as must
evaluations of program effectiveness. Yet it is very difficult to estimate post-
harvest food losses with precision. Partly, this is due to their inherent vari-
ability. But it is also a result of many cultural and economic factors that
frustrate the smooth, efficient flow of food through the postharvest system
from producer to consumer.
Useful food loss estimates are possible, however, as is improvement in food
conservation. This study is devoted to assessing both the potential of food
loss reduction efforts and their limitations. It summarizes existing work and
information about losses of the major food crops and fish; discusses some of
the economic and social factors involved; identifies major areas of need; and
suggests various policy and program options for developing countries and
technical assistance agencies.

Loss Estimation

Any effort to reduce food losses must begin with a quantitative assessment
of the problem. However, loss estimates-unlike production estimates, which
are based on the measurable genetic potential of crops-are location- and
season-specific to a degree that makes the concept of average levels of loss
almost meaningless. The low accuracy of loss-survey techniques on the one
hand, and the limitations of extrapolating from even a specific, well-
characterized loss situation on the other, make reliable, economic loss esti-

*Tonnes (metric tons) are used throughout the report.


mates very difficult to obtain. There is no doubt that losses can be better
understood and assessed, however, and improved methods must be developed
and standardized.
Improved loss estimation is essential for making policy decisions about
the allocation of resources to reduce losses. Experts resist estimating
national or global losses of major food commodities because these figures are
impossible to substantiate statistically, except on a limited, controlled experi-
mental basis. When providing "indicative" figures for planning purposes, the
experts typically cite minimum overall losses of 10 percent for durable crops
(the cereal grains and grain legumes) and 20 percent or higher for nongrain
staples (yams or cassava, for example) and other perishables, including fish.
Even if these estimates are accepted (with appropriate caveats) only as con-
servative minimum values in support of allocations for food loss reduction, it
is clear that worldwide food losses are staggering and that they justify sub-
stantial investment of intellectual and financial resources to better understand
and reduce them. This is reflected in the 1975 Resolution of the VIIth
Special Session of the United Nations General Assembly, committing member
states to reduce postharvest food losses by 50 percent by 1985.

Loss Reduction

The degree of loss reduction achieved will depend ultimately on economic
exigencies. Given modern technology and sufficient resources, it is theoreti-
cally possible to conserve most food commodities almost indefinitely without
loss. The expenditure on food conservation, however, must be justified by
particular needs and circumstances. Before programs can be undertaken to
reduce losses on a national scale, more data are needed on probable costs and
personnel and organizational needs. Actual loss reduction efforts must begin
with political commitment by individual countries to carry through the ac-
tions required at the national level.
Given the complex coordination required to effect loss reduction, each
country requires a national postharvest policy body with a full-time profes-
sional staff to assess and monitor overall losses, identify acute loss priorities,
and carry out research. This body should also provide decision makers with
realistic policy options, so that investment in loss reduction can be made
commensurate with the economic and social costs and benefits involved. The
postharvest policy group must have access to the highest levels of govern-
ment, since losses may result as much from disincentives to conservation
caused by pricing, taxation, or other governmental regulatory policies as from
biological or physical causes.
Regrettably, few countries have postharvest groups responsible for de-
veloping and coordinating policy among the ministries involved. Establish-


ment of such bodies is urgently recommended, and technical assistance agen-
cies should be ready to help developing countries with the process.
Current national efforts for food loss estimation and reduction are not
only inadequate, but are also heavily biased towards storage losses of cereal
grains. Given the seasonal character of grain production and the survival value
of grain in many societies, this is understandable. However, the nongrain
staples, which are the main source of calories, at least in the diets in many
areas, should receive attention commensurate with their importance in the
diet; so should vegetables and fruit. This concern should not, however, be at
the expense of efforts focused on the cereal grains.
In most societies, so much importance is attached to eating fresh, known
varieties of fish (because of the dangers of eating either spoiled or toxic
varieties) that increasing consumption of less-conventional varieties or pro-
cessed fish products (fish flakes or protein concentrate, for instance), is un-
likely to reduce current losses. Rather, efforts should be directed to 1) im-
proving storage; 2) assisting fishermen in forming cooperatives, which
collectively could justify improved boat-landing and fish-handling facilities;
and 3) improving marketing and processing (drying, salting, and smoking) of
landed catches of conventional species. Nonconventional fish species should
be used, wherever possible, for animal feed and fertilizer.

Social, Cultural, and Economic Aspects
Food losses are related as much to social phenomena as to physical and
biological factors. Cultural attitudes and practices form the critical, inescap-
able backdrop for postharvest operations and loss reduction activities. Even
the perception of what constitutes food loss often varies greatly among cul-
tures. The techniques of food conservation are frequently dictated more by
traditional beliefs than by immediate utility. The roles of male and female, or
relationships among individuals and families, may be reflected in the particu-
lar ways in which food is handled or stored after harvest.
Thus, national efforts to reduce food losses cannot rely solely on technol-
ogy or empirical data. Techniques and information must be culturally and
socially acceptable if they are to be useful. Moreover, incentives for the
adoption of sound food conservation practice should be emphasized. Incen-
tives are an important aspect of reducing food losses. Producers are unlikely
to invest money or effort in loss reduction activities unless they foresee a
good return, whether in income, security, or status.
A problem here is that the lack of data about postharvest food losses is
particularly acute with respect to economic and social aspects of loss, mean-
ing that the cost effectiveness of food loss reduction cannot yet be ade-
quately demonstrated. Yet there are simple improvements in conservation
practices that require little monetary investment and could greatly reduce the


risk of serious loss at the farm level. There are also indirect benefits that can
derive from investment in postharvest loss reduction. This is true especially in
the traditional farm sector of poor countries, where the bulk of the popula-
tion produces and consumes the larger proportion of the food crops, little of
which enters the market sector. Here food loss reduction leads to greater
security against lean years. It may also offer possibilities for generating em-
ployment and surplus food for marketing, which may pay for an increased
flow of goods and services to the rural area. Government action to reduce
losses is probably more important in the traditional farm sector than other
sectors of the economy, where food commodities are mainly in the hands of
commercial entrepreneurs who normally respond to market forces with
appropriate conservation measures.

Education and Training

The lack of reliable general information on the extent, nature, and possi-
bilities for reduction of postharvest food losses, combined with lack of recog-
nition that this is a discrete technical area with opportunities for professional
career development, has led to a critical shortage of qualified and experienced
personnel. This should be overcome by educational efforts at many levels.
These efforts should include informal programs to increase public awareness
of the need for hygiene in food handling and storage. They should also
include training courses for agricultural extension workers (who have a par-
ticularly important role to play in the rural farm sector) and administrative
personnel, as well as degree and postgraduate training in appropriate biologi-
cal and engineering disciplines. Particular attention should be given to increas-
ing training opportunities for women, who in many societies play a vital role
in harvest and postharvest activities. Existing technical assistance support for
national postharvest training programs should be strengthened and should be
matched by complementary research and training opportunities in the indus-
trialized countries.

Technical Information and Research

There is little precise published information about losses and loss reduc-
tion in developing countries. That which is available concerns mainly grain
storage; more information is needed about perishables and the socioeconomic
factors affecting food conservation. The literature is scattered widely
throughout the technical journals and is often not readily identifiable by title
as relevant to postharvest losses. There is need for an international post-
harvest loss documentation service, continually updated and with facilities for
providing microfiche or hard copies of technical papers on a worldwide basis.


Although accurate specific data are lacking, a great deal of general technical
and scientific information about various aspects of food loss in industri-
alized countries is available and should be put to use. There is need for adaptive
research to ensure that this information is technically sound and for socio-
economic research to ensure that it is socially acceptable and economically
justifiable to apply it to developing country situations. The private sector in
developing countries is potentially very important because of its information
and experience in the postharvest conservation of commercial and export
Further applied research is needed to improve food processing equipment
so that it will work efficiently under tropical conditions. This applies particu-
larly to drying, threshing, and milling equipment, which is often old, in-
expertly operated machinery designed for other purposes and the cause of
much avoidable loss. There is also a particular need for low-cost, simple
cooling equipment, which could dramatically increase storage and marketing
life of perishables.
Finally, there is a need for basic research, much of it conducted in co-
operation with industrialized countries. Topics for study should include im-
proved, biodegradable pesticides (insecticides, rodenticides, and fungicides) to
be used in integrated systems of pest control, replacing toxic chemicals to
which many pests are becoming resistant and which may be a threat to the
health of people, livestock, and wildlife. The international agricultural crop
research centers and national crop breeding programs should also consider the
postharvest characteristics of new varieties when selecting crops for introduc-
tion to developing countries.
Our study confirms that there is no known simple, inexpensive technology
that can, by itself, make a profound impact on postharvest losses. On the
contrary, postharvest food conservation can be achieved only through a com-
bination of location-specific organization, problem identification, training,
information, and adapted technology. Good conservation practice must be
applied on a sustained basis, with continual refinement in response to new
information. Significant worldwide reductions in food losses will result as the
aggregate of these sustained national efforts, which should be given all possi-
ble support by the bilateral and international technical assistance agencies.

Chapter 1


By the year 2000, it is projected that world population will increase from
4 billion to between 6 and 7 billion. Since estimates indicate that between
450 million and 1 billion people do not have enough to eat now, this number
is likely to increase with the population (NRC, 1977).
To cope with current and future food demand, governments have tradi-
tionally emphasized two lines of action: reducing future demand by slowing
population growth, and augmenting food supplies by expanding production.
A third vital complementary measure, however-reducing the loss of food
during and after harvest-has not been adequately emphasized.
In developing countries enormous losses result from spillage, contamina-
tion, attack by insects, birds, and rodents, and deterioration in storage. Con-
servative estimates* indicate that a minimum of 107 million tonnes of food
were lost in 1976; the amounts lost in cereal grains and legumes alone would
provide more than the annual minimum caloric requirements of 168 million
Billions of dollars have been invested to help developing countries produce
food, but this has not been matched by investment-or by an awareness in
developing countries of the need for it-either to determine what could be
done to reduce losses, or to initiate measures to reduce loss.
Increased food production causes strain on existing methods of handling,
storing, and processing crops, and increased food losses will result unless
developing countries and donors of economic assistance can a) establish and
maintain adequate harvesting, storage, and handling practices, particularly in
rural areas, and b) create efficient policy and administrative infrastructures.
Neither the total magnitude of postharvest food loss nor the extent to
which it is avoidable are reliably known. Losses vary greatly and are a func-
tion of crop variety, pests and pest combinations, climate, the system of
harvesting, processing, storage, handling, and marketing, and the social and
cultural setting. The importance of losses in particular localities varies accord-

*See Chapter 8.


ing to the availability of food and the purchasing power of the various sectors
of society.
Experts involved in the preparation of this report resisted extrapolating
postharvest loss estimates to national or global levels because general esti-
mates cannot be supported with statistically significant data. For planning
purposes, however, 10 percent is cited as an average minimum overall loss
figure for cereal grains and legumes, and about 20 percent as the minimum
for perishables and fish. It is clear that food losses are important to poor
countries in terms not only of quantity, but also of nutritional and eco-
nomic loss.
Many observers believe that a 50-percent reduction in postharvest food
losses in developing countries would greatly reduce, or even eliminate, the
present need of some countries to import large quantities of food. This reduc-
tion has been set by the VIIth Special Session of the United Nations General
Assembly in 1975 as a target to be achieved by 1985. Annual production of
cereals by that time is projected to reach 450 million tonnes, and projected
minimum losses might amount to at least 45 million tonnes, valued at 7.4
billion 1976 U.S. dollars. Calculations for perishables and fish project mini-
mum losses in 1985 valued at over $4 billion for a total food loss in develop-
ing countries valued at more than $11 billion.
Food losses are highly locality-specific, not only in amounts, but also in
their impact: this means they must be evaluated in the context of the relative
economics of food production and of the relationship between food produc-
tion and population growth. There is no doubt, however, of the importance
of loss reduction to governments and technical assistance agencies as a means
of increasing food availability at a time when constraints on production (de-
creased land availability and rising costs of fertilizers and pesticides) are con-
tinually increasing.
Substantial postharvest losses also occur in developed countries. These
losses appear to result from somewhat different causes than those in develop-
ing countries, however. For instance, many stem from consumer demand for
a widely varied diet. Because of strict quality regulations and consumer pref-
erences, a large amount of food is thrown away due to slight changes in
quality or appearance. Similarly, requirements for uniform packaging proce-
dures result in heavy losses of discarded irregularly shaped fruits and vege-

Charge and Objective

The United States Agency for International Development (AID) has identi-
fied postharvest food loss reduction as a key problem area to receive atten-
tion. In order to pinpoint the most appropriate ways to allocate AID funding,


the Agency requested the Academy to undertake a study of postharvest food

The objectives of this study are:
1. To summarize existing work and information on food losses;
2. To discuss some of the social and economic factors involved in food
loss and food conservation; and
3. To identify the need for food loss assessment and food conservation and
to suggest alternatives for food conservation policy and programs for develop-
ing countries and development assistance agencies.

The study does not prescribe conservation projects or practices applicable
to all developing countries, since remedies must depend on each country's
particular circumstances and priorities. Rather, the study reviews alternative
possibilities for reducing losses, presenting them in a way that may help
decision makers to better assess the possible consequences of various courses
of action.
The report is aimed primarily at the decision maker-in both industrialized
and developing countries-who is responsible for resources that might be
allocated to food conservation and who seeks a comprehensive overview of
the postharvest system in developing countries. We have, therefore, included
background and basic technical, socioeconomic, and cultural information.
The report is also designed as a basic introduction for the technical person
not familiar with the field. References and suggested reading are included to
indicate further sources of information.
To initiate and direct the study, the Academy appointed a Steering Com-
mittee whose members have experience with both the technical aspects of
postharvest food conservation in developing countries and the broader scien-
tific, social, and economic context.
The Steering Committee met three times. At the first meeting, participants
agreed on the outline of the study and identified key issues. With these
guidelines established, compilation of a bibliography was begun and informa-
tion solicited from large numbers of experts throughout the world.
For the second meeting, an international group of experts was invited to
join in an examination of the key issues and the roughly assembled study
material. On the basis of these discussions, a final draft was prepared by NAS
staff members for discussion at the third meeting, with subsequent review
under the Academy's report review procedure.

Scope of the Study

Throughout this report, emphasis is given to the major food crops, identi-
fied on the basis of estimates of their levels of production in 1976 (FAO,


1977). The study focused on the basic categories of foods-cereal grains and
grain legumes, nongrain staples, and perishables and fish-in rough proportion
(60:20:20) to their relative importance and the amounts of information be-
lieved to be available about their postharvest problems.
Since Congress has directed AID to devote its main attention to the poor-
est people in developing countries, the study focuses on the needs of rural
farms. Morover, emphasis on the farm sector is logical in terms of produc-
tion patterns; a large portion of all food crops in developing countries remains
on farms and in rural villages and never enters the commercial market.

TABLE 1:1 Major Food Crops, World and Developing Country* Ranked in Order of
Estimated Production (from FAO, 1977)


CROP thousands) PERCENT


Sw. Potatoes

Yams, Taro,
Dry Peas
Dry Beans











Percentage of Total
World Food Crop Production 88.14


CROP thousands) PERCENT

Yams, Taro,

Sw. Potatoes

Dry Beans






Percentage of Total
Developing Country Food
Crop Production







*Developing market economies as defined in the FAO Production Yearbook (1977).
**Pulses-total legumes except soybeans and groundnuts.


As a corollary to the emphasis on crops grown and consumed in the
poorest farm sector, it was agreed to exclude primarily commercial food
crops-the beverages (tea, coffee, cocoa) and other plantation and export
crops such as bananas and sugar cane. These commodities are largely the
province of private enterprise; presumably, the entrepreneurs give postharvest
loss appropriate attention, at least by comparison with the nonmarket food
crop sector.
Meat and dairy products have also been excluded from the study. It was
agreed that they pose special kinds of loss problems related to the provision
of a storage and distribution system. If such a system exists, it operates more
or less efficiently with pasteurization and refrigeration; if it does not, there is
little incentive for production beyond immediate, usually modest, needs and
the products (except for cheese) are consumed quickly with minimal loss.
The study is directed toward losses occurring either in unprocessed food or
in food commodities that have undergone "primary" processing. Primary
processing is a series of steps (taken mainly on the farm, with women taking
much of the responsibility) by which the raw foodstuff is converted into a
basic edible commodity by being treated or separated from inedible constitu-
ents. Rice, for example, is harvested, dried, stored as paddy, hulled and
polished, or parboiled. These steps involve weight loss that may or may not
include food loss, depending on definition. (Rice hulls are not losses because
they are not food, but rice bran may be.)
There are further "secondary" processing steps such as baking, brewing,
or canning in which the basic edible commodity is converted into other forms
before being consumed. The study concentrates on the losses occurring from
harvest through primary processing rather than on secondary processing, a
decision made for two reasons:

Secondary processing takes a large variety of forms. Keeping track of
the commodity as it moves through various stages in this part of the food
chain makes the estimation and quantification of losses a daunting prospect.
In secondary processing, the commodity is normally in the hands of
commercial, village, or domestic processors. Losses are likely to be relatively
small (compared to storage losses, for example), and to the extent that com-
mercial enterprise is responsible, they are probably minimized as much as the
available resources and economic incentives warrant.

Although meat and dairy products are excluded from the study because of
their perishable nature and urgent storage demands, fish is included. This
decision was made because of the importance of fish in the world diet (to
which it supplies 17 percent of animal protein consumed) and because losses
after "harvest" are similar to losses in other perishables resulting from prob-
lems of rapid deterioration, preservation and drying technologies, and storage.


Definitions and Boundaries

Certain key words must be defined to avoid confusion. Perception of loss
is highly subjective and location-specific and the formulation of unambiguous
definitions difficult. The definitions that follow are the consensus of a large
number of knowledgeable individuals who recognize the need for bringing
some uniformity to the use and meaning of commonly used terms. The
definitions are based on those articulated by Bourne (1977).

Food is any commodity produced or harvested to be eaten by a particular
society. It is measured by the weight of edible material-calculated on a
specified moisture basis-that has been harvested, gathered, or caught for
human consumption and that is consumed by the population of the area
under consideration. For the purpose of this study, primary attention is
focused on the major food crops-cereal grains, grain legumes (the "dur-
ables"), and root crops, with secondary consideration given to perishables and

Harvest and Postharvest
Harvest is the single deliberate action to separate the foodstuff (with or
without associated nonedible material) from its growth medium-reaping cer-
eals, picking fruit, lifting fish from water-and all succeeding actions are
defined as postharvest actions.
The postharvest period of time thus begins at separation of the food item
from the medium of immediate growth or production. It is defined here as
ending when the food enters the process of preparation for final consump-
tion. This period also corresponds to the agricultural marketing and distribu-
tion period in which "crop protection" activities have ended, but before meal
preparation activities begin.
Fruit becomes postharvest after it has been picked. Fruit that falls from
the plant and is allowed to rot on the ground is not a postharvest loss because
it was never harvested. However, if fallen fruit is collected for use, it becomes
subject to postharvest loss assessment.

Loss and Damage*
Loss is measured as a reduction in weight in the amount of food avail-
able for consumption. We are concerned here only with losses that could

*For a fuller discussion of this topic, see Chapter 3.


be avoided or reduced given the right conditions under the constraints of
the society in which they occur. Economic considerations may lead to
situations in which it is not desirable to reduce loss that could technically be
avoided. For the purpose of this report, therefore, the food would not be
considered lost.
Damage is physical spoilage, often a partial deterioration or one subjec-
tively judged and very difficult to measure; it is usually reported as a percent-
age of the food sample. Damage of a crop sample is not usually the same as
weight loss and is usually not as useful or precise a loss indicator as percent
weight loss.
Foods that are taboo and therefore not consumed are not held to be lost;
neither are foods used in ceremonial or religious rites. Nonutilization and
underutilization of items not now recognized as food are not considered lost,
though this is an important area of study that should be addressed elsewhere.
It is important that loss definition be location-specific. Cultural differences
create problems in defining loss; what is considered edible, a delicacy even, in
one area (fermented bean curd, for example) may not be viewed as food in
another. Loss definition may even be time-specific, with items rejected in
times of plenty consumed in times of want.

Assessment, Measurement, and Estimation
These terms are used in the literature to describe different kinds of pro-
cesses that determine losses with varying degrees of confidence. They are used
here as follows:
Assessment is used to denote the rough quantitative approximation of
food loss or to characterize the relative importance of different points of
loss in a particular food chain. Implicit in the use of this term is subjective
judgment required because of insufficient information.
Measurement is a more precise and objective process by which quantitative
facts about a loss situation are calculated. Implicit in this process is the belief
that the same procedure applied by any observer under the same circum-
stances will yield the same result. This does not mean that the accuracy of the
result is necessarily higher than that of an assessment-the accuracy will de-
pend on the method of measurement itself, while the accuracy of an assess-
ment can only be borne out by subsequent measurement.
Estimation is used to describe the process of interpretation of a number of
scientific measurements, and thus requires that experience and judgment be
brought to bear on the factual information under consideration.

Waste or wastage are terms included here because they are commonly
used in other reports. However, they cannot be precisely defined since


they involve subjective and even moral value judgments and depend on the
context in which they are used. They should not be used as synonymous with
loss and are probably better avoided.


The need for a survey of bibliographic material was recognized, and this is
included in the study's terms of reference.
As collection of postharvest technology references proceeded, it became
evident that, contrary to expectations, a large amount of material exists that
in some way touches upon loss estimation, food preservation, or storage
This material was organized by major categories: food commodity loss
estimation, conservation technology, and loss vector. The limited time avail-
able and the volume of material precluded extensive cross-referencing, but
country and author indexes are appended.
The present bibliography of some 2,100 entries and 300 pages is recog-
nized to be a working document. Two hundred and fifty copies have been
distributed to institutions actively pursuing research on postharvest technol-
ogy. Additional copies are available through the National Technical Informa-
tion Service (NTIS). The information has also been entered in the FAO
AGRIS computerized information store and can be obtained from AGRIS
catalogs; it is also available from Kansas State University (see p. 173).
To serve the needs of readers who may desire an overview of particular
aspects of the postharvest food loss problem but who have neither the time
for nor the interest in examining large quantities of information of uneven
relevance or quality, selected reading lists have been provided at the end of
the major sections of the report. These lists represent the opinion of experts
on the various topics as to the items in the literature that are informative,
comprehensive, and well-written.


Bourne, M. C. 1977. Post Harvest Food Losses: The Neglected Dimension in Increasing
the World Food Supply. Cornell International Agricultural Mimeograph 53. Cornell
University, Ithaca, New York.
Food and Agriculture Organization of the United Nations. 1977. 1976 FAO Production
Yearbook. Vol. 30. Food and Agriculture Organization of the United Nations, Rome.
National Research Council, Commission on International Relations. 1977. World Food
and Nutrition Study: The Potential Contributions of Research. National Academy of
Sciences, Washington, D.C., p. 1.

Chapter 2

Cultural and

Socioeconomic Aspects

This report emphasizes the technical aspects of postharvest food losses.
Much of the discussion, following the mandate of the study committee, con-
cerns the methodology of loss reduction, the technology of primary food
processing and food storage, and the body of knowledge on food pests and
the physiological deterioration of food. The study committee is fully aware,
however, that prevention of postharvest food losses necessarily involves more
than technical issues. Cultural, economic, and social factors strongly affect
the nature and magnitude of food loss and the attitudes of farm families and
governments to food conservation. Past experiences with agrarian reform have
demonstrated that programs must be sensitive to the cultural, socioeconomic,
and political characteristics of a society and that the technical and scientific
components of change cannot be divorced from the social context within
which they are applied.
The resources available to the committee did not permit a systematic
examination of knowledge about cultural change and the conditions that
facilitate it. This discussion does not represent, therefore, a thorough exami-
nation of the social, economic, and educational issues that need to be con-
sidered in an approach to food conservation. It is intended, rather, to empha-
size the importance of these issues, to point out some of their implications,
and to stress the immediate need for planning to conserve food.

Cultural and Social Factors in Food Conservation

The causes of food loss are linked in many complex ways to beliefs and
attitudes that underlie traditional ways of managing the postharvest system
and that complicate change. These factors must be carefully examined and
understood before new conservation technologies and practices can be suc-
cessfully introduced.


Under traditional farming conditions, the postharvest system of storing
and handling crops is suited to the type, and level of crop production in which
it has evolved, often through a harsh process of natural selection. The levels
of production and conservation of food are constrained by the resources
available to the farm family, and there may be limitations on the time or
labor available on the farm or in the village for incorporating changes in the
established seasonal cycle of events. Furthermore, change may be perceived as
a threat, and resistance to it may be strong. In many societies, for example,
there may be much reluctance on the part of individuals or groups to relin-
quish established controls over food storage and other practices that are
linked to security and status. Traditional practices, therefore, are not likely to
be abandoned unless it can be demonstrated both that new technologies and
methods will be effective improvements and that they will not result in
intolerable strains on social structures, income levels, and distribution.
Despite understandable-and justifiable-conservatism about established
postharvest practices, change is inevitable. Population increase, for example,
may strain food resources and lead to introduction of new crops, new
varieties, or other inputs for production. Such changes will strain the existing
food-handling capacity, creating possibilities for increased levels of loss at all
stages of the postharvest system. Thus, food production increases should go
hand in hand with plans for postharvest conservation techniques and in-
The conditions that foster the incentives necessary to stimulate change
vary over time and among cultures. Past experience suggests, however, that
certain conditions can turn people against technical "improvements" to food
conservation. Typically, these include:

Price depression resulting from increased availability of food;
Taxation, especially tithes on amounts of food stored;
Fixed quotas for commodities to be purchased after harvest;
Inadequate means of storing or marketing surplus production; and
Obstacles to reaching larger markets, such as lack of feeder roads or
inadequate transportation arrangements for fish and perishables.

Conditions that mitigate against food conservation draw attention to the
importance of national policies. Past experience suggests that the effective-
ness of intervention depends on adequate communication between central
governments and local communities. Governments must have adequate in-
formation for planning and decision making. It is for this reason that the
committee places special stress on the need for national policy bodies con-
cerned with postharvest food losses. Such bodies can take account of the
broad range of local interests involved and can examine the technical and
scientific considerations in the light of local conditions and attitudes. This


would aid national governments in deciding how postharvest losses rank in
terms of national priorities.
The decision to act to reduce losses involves not only complex social
considerations, but also economic considerations, some of which are dis-
cussed below.

Economic Factors in Food Conservation

Postharvest losses can arise from a number of causes. These fall into three
main categories, each of which has economic implications:

Physical loss that can be measured by weight;
Loss of quality (including presence of contaminants), with changes in
appearance, taste, or texture that may cause the food to be rejected by
potential buyers; and
Loss of nutritional value.

These losses may affect the subsistence farmer, the farmer who produces
food for sale, and the consumer. Although any losses will ultimately be felt
by society as a whole, individual groups are likely to experience the economic
consequences to different degrees. Further, strategies to prevent or reduce
food losses have economic effects not only on consumers, producers, or
owners, but also on other groups involved in food preservation and pro-
Many kinds of costs may be associated with postharvest losses, and it is
important to assess these costs as thoroughly as possible for an accurate
picture of possible economic consequences.
Individuals or private organizations normally make decisions about dealing
with food losses on the basis of economic consequences alone; governments,
however, are faced with decisions about losses that involve not only economic
consequences but also social responsibility and national development goals.
Clearly, there is no simple, "right" answer for complex and changing situ-
ations, but understanding of the economic consequences of postharvest food
losses can help illuminate feasible answers and eliminate unsuitable ones.
These consequences differ at the production, or farm, level and the broader
social level; both contexts will be discussed below.*

Economic Loss at the Farm Level

For the individual farmer, economic loss is usually expressed in monetary
terms and may result when physical, qualitative, or nutritional loss occurs.

*Discussions on economic losses are based on their treatment by Harman in Adams and
Harman (1977).


For example, a farmer may store grain to sell at a later date; if a portion is
eaten by rodents or is damaged and becomes unsalable, the farmer loses
income he would otherwise have gained. (It should be noted, however, that
the example just given could result in an economic gain where the general
availability of a commodity declines and the price rises as a result; in such a
situation the total income of some individual farmers may be increased.) In a
similar storage loss situation, a subsistence farmer might be forced to buy
extra food to replace his lost supplies and the cost of this food would be a
loss. His diet would also suffer if the food lost nutritional value.
It is also possible for the farmer to avoid an economic loss even though his
commodity has suffered a loss of quality or nutritional value. If such losses
are not detected or the consumer, for whatever reason, is willing to purchase
the commodity at prices unaffected by the qualitative changes, the farmer
experiences no loss of income.
For the farmer, costs related to postharvest food loss may be considered as
direct or indirect and these costs are discussed in Note 2-1.

Economic Loss at the Social Level
Food loss also has implications, of course, for the buyer and the consumer
and thus affects the society or the nation as a whole. (Strictly speaking, the
economic implications spread throughout the entire world, but they are
generally analyzed at the national level.)
For purposes of this discussion, losses at the national level are defined as
social losses. Although the economic implications of postharvest losses will be
considerably more difficult to appraise at the social level than at the farm
level, the causes and consequences of social loss must be recognized.
The difference between postharvest loss consequences for the individual
farmer and for the society as a whole can be shown through examples used
earlier. A farmer may not suffer economic loss, for example, if he sells his
crop at normal prices even though nutritional value has been reduced, but
society incurs a loss through the possibility of poorer health and lower
productivity resulting from nutritionally inferior food.
Conversely, society may benefit when the individual farmer bears an
economic loss. Farmers could, for example, take steps to improve grain
storage that involve substantial costs to each farmer. In the short term, there
could be a surplus of grain and a lowering of prices so that the farmers,
individually and collectively, would lose while consumers benefited. There
may also be effects on secondary groups other than farmers and consumers:
e.g., basketweavers making storage containers may be displaced by the intro-
duction of metal bins.
Social gains or losses also fluctuate in relation to external influences,
notably the world market price for the commodity and the availability of
food from external donors on concessionary terms.


Evaluation of Economic Costs

This report emphasizes the importance of knowing as much as possible
about the actual quantitative or qualitative extent of postharvest food losses
in any given situation in order to make reasonable decisions about corrective
action. The requirement for intelligent decisions is knowledge about the costs
involved in various losses; to the extent practicable, loss situations must be
carefully evaluated in economic terms.

Farm Level Costs
Some illustrations of economic evaluation at the farm level may be
A subsistence farmer may become short of food before the next harvest
and be forced to buy it for sustenance. The money spent or the goods
bartered for the purchased food are a direct cost. If the money to buy food
comes through a loan, the interest paid is also a direct cost. If food has been
damaged in storage and the farmer must sell when the price is low because he
lacks alternative storage, he incurs a direct cost equal to the price he would
normally have received less the price received from the forced sale. In some
instances, a farmer who runs short of food may be helped by donations from
friends or relatives. Although his own direct costs may be negligible as a
result, the cost to the donors must be included in a complete evaluation.
Costs incurred through loss of quality in a commodity may be difficult to
identify. If the loss of quality causes complete market rejection, then the
extent of the loss is reasonably clear. In less clear-cut situations, the analysis
may be aided if the crop happens to be graded, with different prices for
different grades. In the case of quality loss in animal feedstuffs, costs will
vary depending on the value of the substitute feeds used or available.
Nutritional losses due to deterioration of food that is nevertheless still
consumed are difficult to evaluate in economic terms, although it is recog-
nized that they can have an adverse effect on health and productivity. It is
possible to assess the protein and vitamin content of certain harvested crops
in their premium or undamaged condition and assess later decreases against
the premium standard. An evaluation of this nature, however, is essentially
subjective, and interpretations based on such data should be presented sepa-
rately from other aspects of a loss evaluation.

Social Costs
One farmer's postharvest loss will have little social consequence, but the
total of all farmers' losses can represent a significant social cost. Typically,
evaluation of these social costs follows an approach similar to that used for an


individual farmer: the consequences of loss are analyzed as thoroughly as
possible and appraisals made on that basis.
The values assigned to postharvest food losses can be based on the prices
(using international exchange rates) at which the commodity could be traded
by the country concerned. If production of the commodity has been great
enough to meet internal or domestic demand, then a surplus is available, at
least in theory, for export. Losses that occur can be calculated to have cost
the amount of foreign exchange sacrificed by the reduction of exports.
Conversely, if a commodity is not produced in sufficient quantities to
meet domestic demand, some amount of the commodity, in theory, will need
to be imported. Losses can be valued at the cost in foreign exchange of
importing quantities of the commodity equal to the losses.
Just as individual farmers bear indirect costs in coping with or trying to
prevent postharvest losses, indirect costs also can be incurred by society.
Measures taken to prevent food losses, rather than those resulting from actual
losses in a particular season, have indirect costs. The costs of extension staff
who advise on improved postharvest handling and storage of crops are an
example of indirect costs, as would be the costs involved in inspecting and
grading produce to reduce losses. Indirect costs to society can present
problems in economic evaluation, however, because they frequently involve
multipurpose activities and the costs cannot be attributed solely to loss
prevention purposes.

Cost-Effectiveness Analysis
To the extent that prices can be attributed to postharvest food loss
prevention or reduction activities, cost-effectiveness analysis can be a useful
technique for evaluating the cost of reducing a unit of food loss and the
quantity of units that can be affected within a fixed budget. This analytical
approach can indicate which activities could affect the most units within a
fixed level of resources. An example is given in Note 2-2.
While cost-effectiveness analysis can be a useful analytical tool when
reasonably good information is available on costs and anticipated results, it is
only one factor that policy makers and program planners must consider.
Other elements that must be included in the decision-making process are the
sociocultural acceptability of possible programs, overall national development
priorities, and the impact of possible programs on social and economic
matters beyond postharvest food losses. The next section discusses some of
these additional impacts.

A Special Approach to Cost-Benefit Analysis

In a background paper commissioned for this report, Martin Greeley
demonstrates the way in which social cost-benefit analysis supports concen-


tration of loss reduction in the rural traditional part of the postharvest
Greeley points out that differences exist in measuring costs and benefits
for a private entrepreneur and for the public sector for an investment in loss
reduction. The entrepreneur is concerned with private profitability. For
public sector investment, the purely financial considerations are only one
aspect of evaluating the investment.
In the public sector, investment in food loss reduction programs must be
considered in terms beyond the primary objective of loss reduction and
increased food supply. The secondary-but vitally important-objectives may
include effects on employment (including employment of women who are
often displaced by mechanization), income distribution, nutrition, social
stability, and balance of payments. These factors necessarily require that the
financial cost of achieving a given level of postharvest loss reduction be only
one element of evaluation. Poorly conceived programs of loss reduction can
impose social costs that negate the benefits derived from the saving of food.
Well-conceived programs, on the other hand, may not only save food but also
help to provide jobs and distribute income more widely (thus increasing food
purchasing power) as well as saving foreign exchange by reducing food
imports. In such cases, the social benefits can often be of greater weight
than the investment in loss reduction activities.
The postharvest food sector is depicted in the following diagram:


Rural Consumers
Foreign Consumers

As shown, there are three major sources of food and three major types of
consumer. Each arrow represents an element or sector of the total postharvest
The sectors are not always independent in physical and operational terms;
a rural miller, for example, may have as customers both subsistence and
market producers. The sectors do, however, provide a useful division into
target groups for program planning purposes.
According to Greeley, the entire postharvest system has often been neglected
in resource allocations and there is a need to improve operations at all levels.
For planning purposes it is helpful to establish priorities among the six
sectors, which can be defined by the movements of food. These are discussed
in Note 2-3.


The use! of social cost-benefit analysis suggests that, in general, govern-
ments should give greater emphasis to food conservation programs for the
traditional sector. In many countries (especially in Asia), this sector has both
the largest population and the highest total food production, meaning that
with widespread food conservation techniques, substantial savings of food can
be made even when percentage losses are relatively low. Furthermore, the
producers and consumers in this sector comprise the largest poverty-level
group in many countries, and therefore can provide a focus on the poor in
national development activities. Because loss reduction activities may require
some capital investment and must be perceived to provide practical benefits,
governments may need to demonstrate the value of these efforts and to
subsidize them. At the village level, good opportunities exist for using local
raw material, labor, and artisan skills in loss reduction activities; these
elements of a country's resources are relatively cheap and abundant, yet the
opportunity cost-their value in alternative investments-is relatively low. Use
of these rural resources, moreover, provides direct social benefits by generat-
ing employment and distributing income, in addition to reducing food losses
for the benefit of poor farm families and consumers. Increased food avail-
ability in this sector can also provide the poorest farmers and farm women
with access to, and perhaps integration into, the urban and export markets.
Thus, a persuasive case can be made that careful social cost-benefit analysis
will support an increased emphasis on food conservation programs for the
traditional sector. At the same time, it must be recognized that other
demands will frequently have to be given short-term priority. Losses that
affect urban food supplies, for example, are highly visible, often affect local
political elements, and must be addressed with urgency. Other short-term
imperatives can take on equal priority. Moreover, in those countries where the
majority of farmers are involved in market-oriented activities, a focus of
postharvest food loss interventions in the subsistence or traditional sector
assumes less importance.
Despite these cautions, social cost-benefit analysis is an important ana-
lytical methodology for evaluating postharvest food loss reduction activities
in economic terms, whether or not those activities are directed toward the
traditional sector. The results of such activities must be appraised in terms
not only of the reduction in food losses and the costs of achieving those
results, but also in terms of the effects of those savings on the beneficiaries
and of the secondary and lasting impact on the country's overall develop-


Food losses are related to social phenomena, and ways should be found to
incorporate government concern for a country's food supply with the socio-


cultural implications of food loss and food loss prevention. Because food
conservation bears similarities to other types of intervention into rural prac-
tices, the problems and successes of rural health delivery, agricultural exten-
sion, and other community development interventions at the village level
should be studied by those who plan for food loss reduction.
The need to integrate new practices and revised technologies into village
economies calls for a better understanding of traditional practices and,
generally, of the conditions that facilitate or hinder corrective measures.
Thus, there is a need for more research into the links among economic and
cultural practices and food losses.
More understanding is needed of the effects of government financial
policies (subsidies, price controls) on postharvest losses and incentives to
reduce losses. Specific case studies are also needed to illuminate village-level
problems, such as the impact of subsidies on the motivation to adopt new or
changed technology.
There is a particular need for data on the costs of increasing the avail-
ability of food commodities through loss reduction. Such economic evalu-
ation is essential for comparing food loss reduction strategies with other types
of interventions as an aid to more effective decision making and planning of
development programs.


In the examples given in the section, the losses are direct; the farmer has
suffered a decrease in the quantity available for sale, his own consumption, or
barter. In all these cases a monetary value can be applied to the loss. Physical
or qualitative losses may also cause other direct costs to be incurred by the
farmer. A damaged grain crop, for example, may have to be rebagged or
resieved at additional cost.
Indirect economic costs result from measures taken to prevent physical
losses. If a farmer takes steps to prevent future losses, he will normally make
an investment of money and time in the expectation of a positive return. Or a
farmer may prefer to plant an improved crop variety because of its good
yields or marketing characteristics. If he is obliged to produce a different
variety, however, because the improved variety does not hold up as well in
storage, he may suffer some loss in satisfaction, which should be valued if
possible. Indirect costs and losses are more difficult to estimate than direct
losses and should be considered separately when evaluating food loss situ-



In country X, assume, for example, that the national postharvest policy
unit wishes to know how to achieve the maximum amount of loss reduction
for rice, working within a limited budget. Assume further that a program of
introducing new rice milling machinery might reduce annual losses by 600
tonnes, on average, in each province affected, while a particular type of
improvement in village cooperative rice storage facilities might reduce annual
losses by 400 tonnes in each province. If the total budget available for loss-
reduction activities is $2 million, and if the improved storage program costs
$200,000 per provincial project and the milling machinery program costs
$400,000 per province (assume equal rates of amortization), then cost-effec-
tiveness analysis would favor the storage program. This program could be
used in ten provinces, reducing total losses by 4,000 tonnes, while the milling
machinery program could be used in only five provinces, reducing total losses
by 3,000 tonnes, as the table shows.

Cost Effectiveness of Food Loss Reduction Options

Loss Reduction Cost Provinces Affected Program
Program per Province per Province Using Budget* Effectiveness**

Facilities 400 tonnes $200,000 10 4,000
Machinery 600 tonnes $400,000 5 3,000

*Total funds available = $2 million.
**i.e., tonnes of food saved.


The sectors of the postharvest food system in order of importance for
resource allocations are:
1. Subsistence producers to rural consumers. This can be called the
traditional sector, in which the rural consumers are also producers themselves
and labor and service employees are paid in kind. Inputs of capital are
typically very low, and this sector is characterized by on-farm operations of
crop processing and storage.
In many places in the developing world, however, there is an increasing
trend toward market-oriented agricultural production, and many regions or
countries have no identifiable subsistence or rural nonmarket sector in the
strict sense. While small farmers may consume much of what they produce,
they generally also market some portion to meet other requirements, often as
barter. Even when farmers produce one crop entirely for their own use, they
usually produce a second crop for the market. "Traditional," therefore, may


be a more useful description of this socioeconomic situation than "rural
2. Market producers to rural consumers. This sector, called the rural
private market sector, involves large-farmer commercial activities oriented
toward monetary profit rather than subsistence food. More off-farm opera-
tions are involved than in the previous sector, and buying agents, millers and
other processors, and wholesalers and retailers participate in the activity.
Consequently, quantitative and qualitative food losses may be higher because
of the additional transport and handling.
3. Market producers to urban consumers. This domestically produced
urban sector represents the flow of surplus food from the rural production. In
developing countries, distribution activities here are often dominated by
public corporations. Buffer stocks of foods are held within this sector, and
their size is a key variable determining the level of activity for the sector.
Moreover, if the stocks are on occasion too large in relation to management
capability and facilities, losses are likely to increase.
4. Market producers to foreign consumers. The export sector is generally
the smallest of the six divisions and varies from small to nonexistent,
depending on the season and the country. Its importance is in the export of
commodities to generate foreign exchange. The postharvest operations are
generally similar to those of the domestically produced urban sector.
5. and 6. Imports to urban consumers and imports to rural consumers.
Together, these two divisions are called the import sector. Food imports can
be highly variable, and the organization of the required transport and han-
dling facilities for imports may be redundant in times when they are not
necessary. Long-range projection for the mix between domestic and imported
production thus requires high-level policy decisions based on risk calculations
and other factors.


Adams, J. M., and G. W. Harman. 1977. The Evaluation of Losses in Maize Stored on a
Selection of Small Farms in Zambia with Particular Reference to the Development of
Methodology. Report G-109. Tropical Products Institute, London. 149 pp.

Chapter 3

Postharvest Food Loss
and Estimation

In 1975, the United Nations General Assembly, reflecting international
concern with ways to increase the world's food supply, called for a 50-per-
cent reduction in overall food losses by 1985. Progress toward this goal can
be judged only through reasonable quantitative estimates of actual food
losses. Loss estimation is also essential for establishing the programs that will
reduce loss; at the national level, politicians and administrators must have
reasonably accurate information for their decisions on food conservation
Yet, while the committee understands the need for quantitative estimates
to justify budget allocations, we caution against undue emphasis on this
aspect of the problem. Food loss estimation is a complex process yielding
results of limited accuracy. Reliable average figures on losses for a region,
nation, or period of time may be impossible to support with sound statistical
evidence, for reasons to be discussed later in this chapter. Part of the problem
is that standard methodologies for measuring and estimating loss are lacking
for most kinds of food. A variety of estimation techniques do exist for grains,
but considerable care must be taken both in choosing the descriptive termi-
nology and estimation technique appropriate for a given situation and in
using it.
Above all, care must be taken in extrapolating loss estimates from one
situation to another, particularly in attempting to arrive at general national or
global estimates. The dubious accuracy of food loss observations and their
limited general applicability support use of an extrapolated average figure for
loss estimates only under carefully described conditions.
In many cases, it may be unnecessary-or impossible-to make scientific
estimates of loss. The sophistication of measurement required will vary
widely in different situations, and assessment by experienced observers is
often sufficient to justify loss reduction measures. In the commodity sections
of this report, the opinions of qualified observers about average losses are


included; these figures are conservative judgments of the specialists and
should be used with caution.
In many developing countries, the greatest loss assessment need is for a
coordinating body at the national level, provided with an operating arm to
identify where postharvest food losses are occurring and to undertake de-
tailed loss assessment using standard methodologies. This structure will per-
mit continuing review of the resources that should be allocated for reducing
The personnel needed to carry on programs of food loss estimation in
developing countries do not all require a high degree of technical skill. At the
planning and supervisory level, however, it is important that responsible
persons have a thorough grasp of the complexities of postharvest food
processing and distribution, along with sufficient knowledge to call on the
various disciplines essential to loss estimation programs.
Prior to a more specific discussion, there are two general aspects of food
loss estimation that deserve somewhat fuller treatment: the difference be-
tween "damage" and "loss" and the utility, or applicability, of loss estimates.
The distinction between damaged and lost food is often difficult to make.
The subjective term "damage" denotes a condition that is not objectively
measurable. It refers to apparent evidence of deterioration, and its impor-
tance to the consumer depends upon his economic level and cultural
background. A poor family often has no alternative but to consume a certain
amount of damaged food in its diet, whereas more affluent neighbors may be
in a position to exercise selection.
With perishables, damaged portions of root crops, fruit, or vegetables may
be cut off and lost for consumption. However, there will be stages of
deterioration at which the consumer decides that the whole item should be
discarded. It is clearly impossible to define the general conditions under
which a certain type of damage should be considered partial or complete.
This is a culture-dependent decision.
"Loss," on the other hand, denotes disappearance of food and should be
directly measurable in economic, quantitative, qualitative, or nutritional
terms, as follows:

Economic loss is the reduction in monetary value of food as a result of
physical loss.
Quantitative loss involves reduction in weight and, therefore, can be
readily defined and valued.
Qualitative loss, although difficult to assess because it is frequently based
on subjective judgments (like damage), can often be described by comparison
with locally accepted quality standards.
Nutritional and germinative losses, which may be a combination of loss
of quantity or quality, are also difficult to measure.


Loss of food quality through deterioration, contamination, and changes in
the composition of nutrients is important, and needs to be much better
understood and measured. At present, however, quantitative food losses-
from whatever causes-are of more immediate significance, since opportunities
already exist for alleviating the factors responsible for these losses.
As mentioned earlier, quantitative food losses should be determined on the
basis of the food's moisture content. Quantitative estimates of losses can be
used to evaluate the potential of conservation activities in the following ways:

To provide a basis for decisions made by developing-country govern-
ments and international agencies about the allocation of resources for food
production and for postharvest activities such as storage, processing, and
To furnish information necessary for determining the locations and
types of activities that may be effective in reducing losses; and
To increase knowledge and understanding of food supplies.

Relationship Between Accuracy and Usefulness
of Loss Assessment

Despite the limitations inherent in the identification of food losses, prop-
erly selected estimation methods can provide the information essential for
reducing losses. Widespread sampling procedures can be used, for example, in
which untrained observers gather information according to a prescribed
format. Although the accuracy of the individual loss estimates may be low,
large numbers of such observations can provide a useful basis for more general
estimates and for decisions involving extended geographical regions or a
substantial number of food stores.
Large-scale surveys raise the question of how much accuracy is necessary
to make loss estimates that are generally useful. The answer depends upon the
purpose to which the estimates are to be put. With survey procedures, the
range and level of confidence of the individual result is less important than
the overall picture that emerges. If, however, the objective is to determine
losses in specific large-scale food storage or processing facilities and to
institute conservation measures affecting large amounts of food over a num-
ber of seasons, then the accuracy of loss assessment should be as high as
possible. Technologies exist to store almost any food indefinitely, but eco-
nomic, social, and political factors influence the selection of the technology
for a particular commodity and place. Often the type of storage-and the
consequent amount of loss-represents a compromise among factors of stor-
age cost, desired food quality, and the anticipated storage period required.


On another level of assessment, for traditional on-farm storage situations
the degree of accuracy of loss estimates is likely to be low, as are resources
for available corrective measures. Here loss estimation is limited by the
variety and dispersal of storage facilities among families and villages in a given
area and by problems both in sampling procedures and in making generaliza-
tions based upon individual observations.
These problems are likely to be exacerbated by the reluctance of farmers
to provide accurate information and by the efficiency of the traditional storage
methods. Experienced observers agree that in many developing countries
storage losses of grain stored at the farm level are often relatively low,
perhaps on the order of 8-10 percent. If losses at the farm-storage level are
of the same order or smaller than the accuracy achieved by reasonable
estimation procedures, it is obvious that good estimates for a region or a
sector of agriculture cannot be made with any degree of precision by
generalizing from farm data until the number of observations becomes large
and is taken over carefully sampled areas. As the amounts of grain stored
increase, the potential accuracy also increases and there is less need of
subjective judgment.
The limitations of food loss estimation at the farm level lead to the
concept of sound conservation practice, which says that, although it may not
be economically sound or practical to determine precise food losses, certain
food conservation practices are nevertheless justifiable on the basis of com-
mon sense. These could include such efforts as making sure storage bins are
completely cleaned out between seasons, or providing shade and appropriate
containers for transporting and marketing perishables.
Food is such a vital resource in a world of growing population that
reasonable measures to conserve it should be taken even though detailed
information on exact losses may be lacking. Furthermore, although losses at
the individual farm level may be relatively low, in the aggregate the savings
that result from improved food conservation can be considerable.
The complexity of procedures used in estimating food losses should be in
relation to the risk of loss and the quantities of food involved in the situation
under study. Existing data and the opinions of experienced observers may be
useful in formulating a "commodity loss profile" in which loss problems for a
particular commodity are approximated. The approximation can also help
identify areas where losses are higher than would be expected with sound
conservation practices and that therefore require detailed attention.
We have been describing some of the limitations and choices involved in
the estimation of food losses. However, a considerable effort has been made
to develop specific loss estimation methodology and procedures. The remain-
der of this chapter will examine what is known, what is in the process of being
developed, and, finally, what still needs to be done.


Loss Estimation Methodology

The production, processing, and distribution of food involve a system of
movement that is always locality-specific and usually very complex, consist-
ing of many stages. Regardless of the nature of the system, however, certain
food losses always occur.
Often there is a clearly apparent need to estimate food losses. It may be
readily seen, for example, that rodents or insects are attacking stored grain,
and the general extent of loss must be determined to decide whether pesticide
treatment is warranted. In many other situations, however, observation may
indicate that food is being lost in the system, but the quantities, causes, and
specific weak points are not known.
The more that loss estimation is analyzed, the more it is apparent that
there neither is nor can be a simple technique, method, or procedure that can
be universally applied. The movement and storage of commodities between
production and consumption is seldom an easily analyzed flow. Irregular
movement and mixing of various batches in postharvest operations make
sampling procedures and generalizations difficult. Yet, sampling procedures
must be defined precisely according to the particular situation.
Loss estimation in a given situation should be designed so that the
methodology is meaningful, economic, and culturally appropriate. Analysis of
the results should be directly applicable to decisions regarding loss reduction.
It is important, therefore, to integrate the process of reducing losses with the
process of loss assessment. At the farm level, for example, the limited
resources available for estimation should also be applied to reduction to be
credible to the farmer; estimation must not be seen as an end in itself.
For a variety of reasons, more techniques have evolved for the estimation
of grain losses than for other major food categories. These reasons will be
examined in greater detail in subsequent chapters on specific commodities;
suffice it to say here that these techniques reflect the importance of grains as
staple foods, the relative physical uniformity of specific grains, and the ease
of storing grain.
Grain loss estimation methodology has recently been the subject of a
manual prepared by Harris and Lindblad (1978) for the American Association
of Cereal Chemists and the League for International Food Education, sup-
ported by funds from the Agency for International Development (AID). The
manual is designed to be widely used in developing countries to encourage
standarized loss assessment procedures so that results from observations
carried out in different locations can be more easily compared. This valuable
document has been written in consultation with grain loss experts involved in
the major national and international programs around the world, many of
whom were also involved in the preparation of the present study. Thus, these
two AID-supported projects are complementary-the manual designed for


those directly involved in grain loss estimation, and this study designed to
cover a wider range of subject matter for a more general audience.
A number of useful techniques and approaches have been developed for
handling problems of food loss estimation. These techniques can be applied in
situations ranging from a broad analysis of where losses occur and at what
rough levels of magnitude to sampling and estimation procedures yielding
rather precise loss figures. The techniques, described in the following para-
graphs, include a) overall assessment of the commodity movement system, b)
field investigation of losses, and c) loss measurement (or "experimental
The following pages describe the major methodological aspects of loss
assessment, including inherent problems and limitations and future needs.
Those seeking more detailed information on procedures for grains should
consult the Harris-Lindblad manual directly.

Overall Assessment

Overall assessment of the commodity movement system means a search for
the points where the most acute food loss occurs. It implies study of the
whole physical and social system in which the food moves from producer to
consumer, and will identify how the commodities are handled (size, number
of steps, etc.) and the number of participating middlemen. Its objective is to
permit judgments to be made about the possibilities for loss reduction
interventions. From the loss assessment and reduction perspective, it will be
helpful if a national policy body exists to deal with postharvest loss problems,
to coordinate the efforts of national and international assistance agencies, and
to gather and analyze loss information. Relevant loss information can be
obtained from a variety of sources: ministries of agriculture, central statistics
organizations, university faculties of agriculture and economics, transporta-
tion agencies, marketing boards, commercial organizations, and farmers' co-
Locality- and commodity-specific information is needed to develop a
"commodity loss profile" describing the movement of a commodity through
the system and highlighting points of potential or actual food loss.
The figure below depicts the "food pipeline" and the physical and bio-
logical ways in which some losses occur. It must be emphasized, however,
that the actual movement of food from harvest to consumer may be simpler,
or may involve a much more complex system than that represented. Move-
ment can be irregular or can be halted for long periods of time; batches of a
commodity can be divided and routed through the system by very different
paths and schedules; infusions of a commodity into the system can be made
from different sources.


The "pipeline" also has a number of different kinds of materials. There are
the human and the mechanical parts of the pipeline, the chain of hands and
the line of the transport vehicles through which food passes with greater or
lesser efficiency, speed, and ease. The food in the pipeline is propelled by
socioeconomic and political forces; regulations and other bureaucratic pro-
cedures slow down or accelerate the food's passage from producer to con-


* Rain
a Humidity

* Broken Spillage
Grain Bruising
* Excessive Breakage
Dehulling Leakage
* Trimming


E Excessive

[Adapted from Bourne, 1977]

Despite the complexities of the system of commodity movement, experi-
enced professionals can make useful estimates of losses and identify possibil-
ities for loss reduction. Simple observation, for instance, of such visual
indexes as insects, mold, or leaking roofs may be all that is necessary.
Further, such factors as the use of pesticides or the type of storage facility
can provide a knowledgeable person with a basis for judging where losses
occur and in what magnitude.
The ultimate use of a commodity also bears on loss estimation. Harvested
grain may be divided into several lots for different purposes, with each
receiving different treatment-some dried and stored for long periods as seed
and some held only for short-term storage and consumption or movement off
the farm. Different loss risks would be involved for the different uses; for
instance, farmers frequently consume their low-quality grain first, since it is
known to be subject to the most rapid loss.


These observations enable the trained observer to develop a commodity
loss profile for a particular commodity. Such a profile would indicate the
final uses of the commodity, the channels through which it travels to final
use, the points at which losses occur, and rough estimates of the relative
magnitude of the losses at these points. It should be mentioned that complete
information on food handling frequently is not collected, but the data is
critical, i.e., the number of handling steps involved, the number of middlemen
handling the food (with inevitable losses) at each step. It is only this kind of
complete information that will enable the expert to judge with confidence
what should be investigated and where priorities are to be assigned.

Field Investigation of Losses
Field investigation typically results from analysis of critical points of
potential or actual loss in the commodity loss profile.
The first step is to develop methodology for the particular objectives of
the investigation. Loss assessment projects frequently have suffered from
poorly defined objectives and from lack of experimental control. In addition
to objectives, the project must have a pattern that is replicable so that loss
comparisons can be made. Comparisons must be statistically valid and must
be undertaken within a logical framework of field investigation and scientific
There are two aspects to field investigation: the survey and the sampling
First, a survey is made of farms, villages, or areas to determine the
locations at which the loss assessment will take place, the parts of the
postharvest system to be investigated, and the farms or villages from which
samples will be taken. A recognized statistical procedure should be employed
for selecting farms or villages if it is intended to apply the resultant loss data
to an estimate of loss over the area as a whole. Adams and Harman (1977),
for example, recommended stratified random sampling in connection with
their assessment of losses of maize in Zambia.
The method of sampling a commodity is the way in which the sample is
removed from the location under investigation, such as a farm or village store.
Specifications of sampling procedures are frequently missing from the post-
harvest literature, yet are a critical dimension of loss estimation procedures.
If the purpose of sampling is to estimate the loss in all the produce in a
store at a particular time-for example, during one or two visits in the
season-then the sampling must be carried out on all the produce in the store.
If sampling is undertaken at regular intervals over a season, on the other hand,
then each sample should be taken only from produce being consumed
between samplings; to remove other samples would disturb the natural
process of loss.


The size of a sample is limited by practical considerations, including
whether or not the sample is being removed for analysis and return to the
store. (With maize, for example, Adams and Harman (1977) suggest samples
of 10 cobs or 1 kg as a reasonable quantity.)
Sampling of stored grain also must take into consideration removal of the
commodity from the store for normal consumption or sale. Large losses
quoted in the literature often reflect heavy damage to a small amount of
residual stored commodity at the end of a season, while in fact total weight
loss of the entire crop may be much smaller.

Estimation of Total Losses

After calculating losses in commodity samples, the investigator still faces
the problem of estimating total loss in the entire lot under scrutiny. In
making such estimates, it is important to relate losses to the pattern of
consumption. If, for example, food is left untouched throughout the storage

Sampling stored grain, People's Republic of China (Courtesy E. S. Ayensu)


period and at the time of removal the estimated loss is 10 percent, then this
represents the total loss over that period. However, in most cases food is
removed at intervals during the storage period, and each quantity removed
will have suffered a different degree of loss since it will have been exposed to
deterioration for a different length of time. The total loss over the season can
be obtained by accurately weighing all the grain in and out of the store and
comparing the totals. This does not, however, indicate the relationship
between loss and time; that is, whether the loss reached a peak or whether it
was related to a particular part of the season.
Clear distinction, obviously, must be drawn between observations of loss
made at different stages of the system, whether they are made on the same
lot of grain experiencing all the losses cumulatively, on different lots, or on a
mixture, as grain is added or withdrawn (see Note 3-1). Sophisticated meth-
ods are available to deal with these kinds of loss estimation problems for
grains (Harris and Lindblad, 1978); they do not yet exist for perishables. The
whole area of deterioration over time of stored perishables in developing
countries, and the implications for costs of loss reduction, need priority

Interpretation of Results
As Adams (1976) points out, it is clearly impossible to avoid approxi-
mation in estimating storage losses of subsistence farmers unless enumerators
can be used within each village to check and weigh each removal of stored
grain. In most cases, provided the same method of estimation and similar
approximations are used for a well-selected sample, the loss estimates will be
comparable and will enable decisions about loss reduction activities. The
pattern of loss and the factors influencing it should also be recognizable. If
possible, loss reduction activities are to be evaluated effectively using accurate
weighing of food quantities, replication, and simulation of normal usage. The
data for this evaluation should cover the whole storage season and can best be
obtained from the type of general loss survey described in the previous
Food loss estimation in developing countries is plagued by the inverse
relationship between accuracy and extrapolation. At one end of the scale, the
few trained observers who have the time, experience, and the trust and
cooperation of the farmers whose losses are being estimated can obtain results
of reasonable accuracy, but with limited extrapolation to other situations. At
the other extreme, large numbers of poorly instructed, untrained observers
are likely to produce representative information of low accuracy and little
The inherent variability in postharvest food losses renders extrapolation of
estimates from one loss situation or from one time period to another


difficult, if not impossible, without being so misleading as to be counter-
productive. At present, available information is so limited, even in the case of
the cereal grains that have received most attention, that experienced observers
agree it will not substantiate the use of single "average" or "representative"
values for losses of food commodities at national, regional, or global levels.
The available values for losses in particular situations should be used only as
indicative of the particular kinds of losses in highly similar situations.
Until much more research and loss assessment is undertaken by trained
observers on a planned, systematic basis using well-conceived standard
methodologies, aggregate estimates of loss that can be substantiated by
statistically sound observations will not be possible. In the interim, where
these values are required for planning purposes the conservative judgment of
experienced observers familiar with the local situation is the only basis for
arriving at a particular figure. This figure is apt to be meaningful to the extent
that it can be related to particular situations and backed up by experimental
data. Again, it must be emphasized that there is great danger that these "best
judgment" figures will be taken out of context and quoted as authoritative, as
has so often happened in the past.
The committee is persuaded that the magnitude of losses of all commod-
ities justifies additional efforts to gain knowledge about their nature and
extent, thus providing an information base that can lead to improved con-
servation of food. This is particularly so for losses of perishables, about which
there is an almost total lack of reliable scientific data.
The committee, therefore, recommends that additional resources be allo-
cated by developing country governments and technical assistance agencies to
improving knowledge of a) the movement of food from source to consumer,
and b) the locus, nature, and extent of postharvest food losses. Priorities for
these efforts should be assigned in proportion to the importance of the food
in the local diet. This should not be interpreted to mean that the committee
recommends the diversion of resources from existing efforts directed at cereal
grains and legumes to efforts concerning other crops but, rather, the commit-
tee's recommendation aims at redressing the imbalance of effort that so far
has been aimed primarily at the durable crops.
The critical shortage of trained observers for identifying and estimating
food losses should be alleviated by appropriate short- and long-term training
programs. Innovative approaches, such as use of rural high school and univer-
sity students as observers under appropriate supervision, should be considered
by the responsible national body as valuable supplementary resources for loss
estimation and reduction studies. One source of expert skill sometimes found
in a country lies within the marketing organization for valuable export crops
such as cocoa, copra, coffee, etc. Diversion of some of these valuable skills to
the subsistence economy should be encouraged.


Additional information on methodologies of estimating grain loss resulting
from particular causes is given in Note 3-2.
Postharvest loss estimation methodology for perishables (including fish) is
much less advanced than for cereal grains and legumes. The development of a
standard methodology is complicated by a number of factors intrinsic to the
nature of the commodities. Differences such as the following must be taken
into account:

The high moisture content of the harvested perishable material, which
makes estimation of weight loss on a dry matter or defined moisture-content
basis difficult, if not impossible.
The lack of uniformity in weight and shape of individual perishable
food items as compared with rice, wheat, or other grains and legumes.
The potential in perishables for partial loss. In the case of fruits and
vegetables, the size of the food and its susceptibility to deterioration at
different rates in different parts make it possible to divide the edible parts of
an individual yam or banana, for instance, into acceptable and unacceptable
portions; grains, by and large, are either edible or not.
The rate and consequence of spoilage in perishables, which in fish is
particularly rapid and potentially dangerous to the health of the consumer,
but is also important in other perishables.
Differences in the relative value of each food unit, individual perishable
items being more valuable than individual grains. The value is not only
economic, but nutritional, particularly in the case of fish.

At present, all that can be reasonably recommended to bring some order
to the estimation of postharvest losses of perishables-a new field-is that
research workers be as explicit as possible when reporting what they are
measuring. For example:

For roots and tubers, weights should indicate whether the observations
were made on fresh, cured, or aged material; whether with or without skins;
and whether vegetative reproductive parts have or have not been removed.
For fruits and vegetables, weights should specify whether observations
were made on fresh, whole material or whether skins, peels, cores, etc.,
were removed.
For fish, the situation is complicated to an extraordinary degree by the
unique characteristics of the "harvesting" process and the many ways in
which the fish harvest can be measured. The differences between total-catch
live weight and landed catch on the one hand, and the weight of edible
portions of individual live, gutted and deheaded, or dried fish on the other,
must be clearly distinguished in reporting and discussing loss.


With the nongrain staples, there is particular need for case studies of
different crops and situations to develop methodology for estimating losses of
commodities on a standardized basis. Establishment of locality-specific stand-
ards of quality for perishables is also urgently needed.



For example, rice loss estimates for Southeast Asia are reported (De
Padua, 1974) as follows:

Harvesting 1- 3 percent
Handling 2- 7 percent
Threshing 2- 6 percent
Drying 1- 5 percent
Storing 2- 6 percent
Milling 2-10 percent

The possible range of weights of food lost as the grain passes through these
stages is not the same as the simple sum of the percentages of loss, since the
weight of a given lot of grain is reduced at each stage. Assuming that there is
no removal of grain other than through loss and no dilution of the lot by
addition of grain, the sum of losses in the example given above would be
calculated as follows for a 100-kg lot of paddy:

Loss Grain In Grain Out
Stage Percentage (kg) (kg)

Harvesting 1-3 100 97-99
Handling 2-7 97-99 90.21-97.02
Threshing 2-6 90.21-97.02 84.80-95.08
Drying 1-5 84.80-95.08 80.56-94.13
Storing 2-6 80.56-94.13 75.73-92.25
Milling 2-10 75.73-92.25 68.16-90.41

Where there is withdrawal of grain at any of the stages, or dilution of the
original lot with added grain at any stage, appropriate adjustments in the
observations and calculations must be made.


Specific Crop Loss Assessment Considerations

The following paragraphs outline the procedures involved in the loss
assessment of different cereal and legume commodities. It should be kept in


mind that there is a need to distinguish between crops that are cultivated
and place a major demand on available labor, such as rice, as opposed to those
grown as a minor activity. The differentiation is significant in determining the
relative importance of the crop in terms of postharvest loss reduction. The
input labor cost is a decisive overall factor in which, in particular, the role of
women on the farm is important and frequently overlooked.
On the basis of major review papers, original published material, discus-
sions with experts, and firsthand field and laboratory experience, Harris and
Lindblad (1978) conclude with regard to techniques for measuring cereal
grain losses:
All of the U.S. Food and Drug Administration-generated procedures,
which are employed as standard methods of loss measurement in the USA
and other countries, are too time-consuming, require a laboratory setting,
require judgements that are difficult to standardize, use sample sizes that
are too small, or have too variable a relationship to grain weight loss to
make them suitable for use in developing countries.
This results from the very different conditions under which losses occur
and are estimated in developed, as opposed to developing, countries. The
same sequence of food movement, storage, and marketing occurs in both
kinds of countries: the degree of loss at different stages differs markedly and
the commodity loss profiles are different.
In the highly mechanized agricultural production typical of developed
countries, losses, generally speaking, are proportionally greater during the
harvesting process; smaller during processing, storage, and handling; and great-
er again during marketing; consumers demand variety and high quality of
products, and this, together with government standards and regulations,
leads to high "shelf" losses at the market level as well as at the table.
In developing countries, on the other hand, while many components of the
commodity loss profile are similar, losses tend to be low during harvest,
where the crop will be mainly handpicked, high where processing involves
primitive procedures (threshing grain with animals), high during storage, and
somewhat lower after marketing.
Thus, the FDA-developed tests (and by analogy other equivalent devel-
oped-country procedures) are largely designed for completely different condi-
tions, that is, for monitoring large-scale modern storage rather than for use
under field conditions, and may require expensive laboratory-based apparatus
and procedures that are time-consuming and difficult to standardize.
Discussion follows of grain losses due to insects, fungi, vertebrate pests,
and handling and primary processing.

Losses Due to Insects
Examinations for insects on the surface of the grain, weighing insect frass,
and various procedures to detect visually damaged grains and count or weigh


them have been given field trials in developing countries. There is a positive
correlation between damage, insects, and frass with some loss quantification
possible and Harris' (1972) report to the World Bank suggests their use in
making rapid assessments. This information may be useful in developing a
commodity loss profile and in quick comparisons to identify likely points of
acute loss in the system.
Some confusion exists concerning the application of these procedures in
quantifying actual losses. Their use in test situations and positive correlations
to weight losses have been taken by some to indicate that they can be used
with some degree of precision to determine weight losses. In fact, they cannot
be used for this purpose unless the biological and physical characteristics of
each estimation situation are completely understood.
All of the procedures, however, are of value in assessing a situation and
coming to a personal judgment. Their precision as indicators of actual losses
depends upon the expertise of the user. This is also true of the so-called
gravimetric techniques in which a comparison is made between the actual
weight of a sample and the weight it would have had in the absence of
For reliable testing, then, Harris concludes that loss in weight can only be
determined by comparison weighing such as "before" and "after," equal
volumes "with" and "without" treatment, and other methods for various
causes of loss as summarized for cereal grains by Adams and incorporated
into the manual prepared by Harris and Lindblad (1978).
However, although comparison weighing is necessary for weight-loss esti-
mates, it is more complicated than simply weighing appropriate samples at
successive intervals on a balance of appropriate accuracy because the moisture
content varies throughout the year. The following paragraphs will serve to
illustrate the complexity of the procedure required to compensate for
changes in moisture content, without repeating the excellent technical treat-
ment given in the Harris-Lindblad manual.

Losses of grain prior to secondary processing are mainly due to insects
and molds. The insects bore into the kernel and feed on the surfaces,
removing food (sometimes selectively), permitting increased uptake of
moisture by the grain from the atmosphere and encouraging the growth of
There are two ways in which the weight loss can be measured: by
weighing a measured volume of grain-in which case the change in weight
in successive samples tested over a period of time is a measure of their
losses (and possibly other factors-the cause has to be determined)-or by
separating damaged from sound kernels in a given volume and measuring
their comparative weights calculated in terms of the whole sample.
In the first case it is necessary to express the weights of grain in terms
of a constant moisture content-usually the dry weight. While it is rela-
tively simple to measure moisture content with a moisture meter, the


volume of the grain changes slightly with changes in moisture content, so
it is necessary to measure, by experiment, the weight of the standard
volume of grain at different levels of moisture. Then the weight of sub-
sequent samples taken at the prevailing moisture content can be corrected
to the original moisture content for weight change to be calculated inde-
pendent of moisture.
Because different varieties of grain have different characteristics, graphs
of weights of the standard volume of grain at different moisture levels are
necessary for each variety. The procedure also is based on the assumption
that the grain is homogeneous; if it is not, as in the case of lots of grain of
mixed varieties, then a separate baseline graph is required for each lot.
The baseline graph is prepared by taking a bulk sample of approximate-
ly 5 kg from each store under consideration (or for each variety). The bulk
sample is sieved and its moisture content measured; it is then divided into
five sub-samples to correspond to five points on the weight/moisture con-
tent graph. Since the normal range of moisture content in stored grain is
from, say, 8 to 18 percent, it is necessary to select five points within that
range. The original 5-kg sample will have a moisture content somewhere
between these two extremes; the other four sub-samples must be dried
down or wetted up to the selected percentages to complete the required
range, which in itself is a procedure requiring care and time.
Three replicate standard volumes of grain taken with a chondrometer
(the test weight container) according to the instructions provided by the
supplier are then weighed to the nearest 0.1 g and the mean weights of the
five samples converted to dry weight plotted on a graph against moisture
content. This graph can then be used throughout the sampling period to
measure the dry weight of samples at any moisture content. The sample
whose weight is to be measured is sieved (the weight of sievings being
considered as losses if they are not used as food or calculated back to the
weight/volume if they are), its moisture content is measured, and
sub-samples are taken three times with the chondrometer. The samples are
weighed, dry weight is calculated from the graph, and the weight change
calculated by comparison with the original sample at the beginning of the
test period.
Error is introduced by factors affecting variation in packing and, hence,
the volume of the chondrometer-sampled grain. These factors include high
levels of damage, which increases packing, and presence of insecticide dust,
which reduces it, so that treated and untreated grain should not be com-
The comparative weights of undamaged and damaged kernels gives per-
centage weight loss directly on the assumption that the undamaged por-
tion is completely undamaged. The disadvantages of this method become
apparent at high and low levels of damage: hidden infestation results in
underestimation of loss, because grains that have lost weight are included
in the undamaged group. At high levels of damage it may be difficult to
identify and count damaged grains accurately among the debris. It also
assumes that insects choose grains at random, which for maize is not the
case. Nevertheless, particularly for unshelled and mold-damaged grains, it
provides a useful means for estimating loss at moderate levels of infesta-
tion with a minimum of equipment.
Other methods are variations of these procedures. The chondrometer
method is the preferred method as it has the highest accuracy when prop-


early carried out. It is neither simple nor rapid, however, and requires a
fairly high level of experience, if not of training, and a variety of ancillary
equipment not readily assembled outside a laboratory.

Losses Caused by Fungi
The methods for estimating loss from insects are also applicable to fungal
damage. However, when mold occurs, a considerable proportion of the grain
rejected by the farmer is often discarded or used to feed animals, since the
presence of infected grain causes a drop in quality grading. The impact of
fungal infection on loss can be estimated by including the separation of mold
damage from other types of damage during the analysis.
The quantification of "weight loss" when the loss is due to fungal damage
will depend on local practices in the use of the damaged material. People
accept or reject damaged kernels as local custom and hunger dictate. It is
desirable to make measurements in one country, or region of it, that can be
compared with measurements made elsewhere; in each situation, acceptance-
rejection limits should be defined in terms of a widely used language. Despite
the difficulty, these limits, based on information from interviews, must be
It seems likely that methodology for fungal damage estimation will need
to be somewhat separate from that for insect loss, but since the two are
frequently interrelated and interacting, the degree of separation needed is
currently unclear, and likely will be situation specific. In sampling, allowance
must be made for differences in moisture content of infected and uninfected
The hidden or socioeconomic effects of moldy grain that may be con-
sumed are more difficult to assess; they are related to the possible presence of
toxins and the tendency for repeated consumption of infected grain to cause
chronic illnesses. This will lead to a reduction in work output by an affected
person and may be likened to the effects of nutritional loss.
Losses caused by fungal contamination can arise through:

1. The rejection of food because of visible fungal contamination or fungal
2. The rejection of food (which may well not be visibly contaminated
with mold) because of its mycotoxin content.
The mycotoxin contamination of food can arise from:
a. The direct fungal contamination of the food.
b. The consumption of mycotoxin-contaminated feed by animals lead-
ing to contaminated animal products (e.g., meat and milk). Rejected food is
often used as animal feed.
3. A decrease in the yield of food.


The ingestion of contaminated feed can reduce the productivity of
animals (e.g., a decrease in milk yield). Sufficiently high doses of mycotoxin
can result in death.
4. Acute and chronic illness caused in humans by the ingestion of con-
taminated food.

Because of the increasing awareness of the mycotoxin (especially aflatoxin)
problem, there is a corresponding likelihood of increased food rejection. A
rapid method exists for observing aflatoxin in maize (utilizing the BGY
fluorescence) and groundnuts are routinely sorted using electronic color
sorters. Established assay procedures also exist for the analysis of a wide
range of foods and feeds for aflatoxin and other mycotoxins.
Recent examples of food rejection, after analysis, include the rejection of
large quantities of corn in Zambia and of shipments of wheat in Pakistan and
In some West African countries the groundnut crop contributes greatly to
the GNP of the country, and therefore food and feed losses can represent
severe economic losses.

Losses Caused by Vertebrate Pests
Losses caused by vertebrates such as rodents and birds are difficult to
assess directly, since they result in the removal of grains from the store. The
usual method of estimation is to blame vertebrate pests for all losses that
cannot be accounted for in any other way. It is difficult to obtain an accurate
estimate without accurate weighing of the grain throughout the season.
Another method is based upon an estimate of the pest population, usually
by trapping and then conducting consumption trials with captured animals to
obtain a figure of daily food intake. However, allowances need to be made for
situations in which the store is not the only source of food; one must also
account for the difference between unlimited food supply in the consump-
tion trial and the foraging required in the field situation.
Studies carried out under warehouse and village conditions have shown
that rodent populations can probably best be estimated by combining a
number of techniques, including rodent sign survey, trap-release-trap, and
estimating consumption of poison baits. In captivity the roof rat (Rattus
rattus) has been found to consume 8-12 g/day of food grain; the house
mouse (Mus musculus), 3-5 g/day; and the bandicoot rat(Bandicota spp.),
25-30 g/day. While eating, the rats also contaminate an estimated 10 times
more food than they eat with urine, feces, hair, and saliva (Yashoda et al.,
These estimates are difficult to extrapolate with confidence because they
neglect the fact that rodents often hoard amounts of food many times greater


than they actually consume. Thus, predictions of losses from rodent popula-
tion estimates are likely to underestimate actual food losses (Frantz, 1972).
Postharvest losses due to rodents have been summarized by Hopf et al.
(1976). Estimates of damage are quite variable and range from 0.5 percent to
60 percent. Amounts are given in some cases, with India reporting approxi-
mately 11 million tonnes lost annually with a value estimated at over $1,000
Postharvest losses due to birds are likely to be of relatively minor impor-
tance compared to preharvest losses to birds and postharvest losses to other
factors. Exceptions occur where grain is left in the field after harvest or
spread in the open to dry for long periods, or where stores allow birds access.
Other than figures of weighing before and after, including all sources of loss,
there is little information on postharvest loss to birds per se. Guggenheim
(1977) reports distinguishing bird damage from insect and rodent damage in
millet on the cob by observation of the marks on the ear.

Losses Due to Handling and Primary Processing
These are losses that may occur at the following stages of the grain
postharvest system:

bagging, or placing threshed grain in other containers
transport from field to storage
transport to mill
milling, which may involve several processes and stages
transport from mill to storage or market.

These losses should be determined by weighing before and after the
particular step or by weighing the amount of grain or grain products in food
and nonfood categories. In many cases, appropriate methodology will have to
be developed to meet particular local handling procedures, as grain is moved
from field to store, mill, and home by different methods. Steps in the pro-
cess are points of potential loss to be investigated.
There is very little published information on postharvest grain losses
during transportation. Yet, any transfer of grain from one stage to another
implies the possibility of loss.
There are three ways in which transportation losses may particularly

1. During handling of crops between harvest, threshing, storage, and


2. In connection with transportation allied to storage-"moving storage"-
during which loss may occur, for example, due to continued deterioration of
bagged grain in transit; to spoilage of bagged grain exposed to rain during
transportation; or to spillage due to container damage or inefficient transfer-
ence of the grain. The use of hooks to handle sacks of grain in port facilities is
a frequent glaring example of loss caused by handling during transportation,
but the use of old sacks from which grain leaks, and which permit access by
pests, is perhaps even more important.
3. As a result of the absence or inefficiency of transportation facilities
and of limited access to alternative market possibilities.

As we concluded earlier, it is probably not productive to pursue maldis-
tribution as a source of postharvest loss of grain, since this involves many
nontechnical factors whose effects reach beyond the postharvest environ-
ment. Maldistribution is more relevant to postharvest losses of perishables due
to the overriding importance of moving the commodity to market as quickly
as possible after harvest.
The limited information available about these aspects of transporting food
provides little specific wisdom regarding food loss. But it is obvious that
attention should be given to transportation problems in the interest of overall
security and efficiency of delivery of food supplies.

Recommended Reading

Adams, J. M. 1977. A review of the literature concerning losses in stored cereals and
pulses published since 1964. Tropical Science 19(1): 1-27.
Adams, J.M., and G.W. Harman. 1977. The Evaluation of Losses in Maize Stored on a
Selection of Small Farms in Zambia with Particular Reference to the Development of
Methodology. Report G 109. Tropical Products Institute, London. 149 pp.
Asian Productivity Organization. 1974. Training Manual: Postharvest Prevention of
Waste and Loss of Food Grains. APO Project TRC/1X673, Asian Productivity Orga-
nization, Tokyo. 358 pp.
Brown, R.Z., and H.D. Pratt. 1976. Biological Factors in Domestic Rodent Control. Rev.
ed. U.S. Public Health Service, Center for Disease Control, Atlanta, Georgia. 32 pp.
68 refs.
Christensen, C. M. 1974. Storage of Cereal Grains and Their Products. American Associ-
ation of Cereal Chemists, St. Paul, Minnesota. 549 pp.
Christensen, C. M., and H. H. Kaufmann. 1969. Grain Storage: The Role of Fungi in
Quality Loss. University of Minnesota Press, Minneapolis, Minnesota. 153 pp.
Food and Feed Grain Institute, Kansas State University. 1976. Grain storage and
marketing short course outlines (in English-Spanish-French). A. Fundamentals.
B. Grain inspection and grading. C. Handling, conditioning and storage. D. Sanitation.
E. Marketing, operations and management. Mimeographed. Department of Grain
Science and Industry, Kansas State University, Manhattan, Kansas.
Greiffenstein, A.C., and H.B. Pfost. 1974. Moisture Absorption of Bulk Stored Grain
under Tropical Conditions. Research Report No. 6. Food and Feed Grain Institute,
Kansas State University, Manhattan, Kansas. 76pp.


Hall, D. W. 1970. Handling and Storage of Food Grains in Tropical and Subtropical
Areas. Agricultural Development Paper No. 90. Food and Agriculture Organization of
the United Nations, Rome. 350 pp.
Harris, K. L., and C. Lindblad. 1978. Post-Harvest Grain Loss Assessment Methods.
American Association of Cereal Chemists, St. Paul, Minnesota.
Munro, J. W. 1966. Pests of Stored Products. Hutchinson & Co., Ltd., London. 234 pp.
Pedersen, J.R., R.B. Mills, G.J. Partida; and D.A. Wilbur. 1974. Manual of Grain and
Cereal Product Insects and Their Control. Department of Grain Science and Industry,
Kansas State University, Manhattan, Kansas.
Phillips, R., and S.G. Unger. 1973. Building Viable Food Chains in the Developing
Countries. Special Report No. 1. Food and Feed Grain Institute, Kansas State Uni-
versity, Manhattan, Kansas.lOlpp.
Pingal, S. V.; K. Krishnamurthy; and T. Ramasivan. 1967. Rats. Food Grain Technol-
ogists' Research Association of India, Hapur (U.P.), India. 91 pp.
Ramirez-Genel, M. 1974. Almacenamiento y Conservacion de Granos y Semillas. 2a
impression. Compania Editorial Continental, S.A., Mexico City, Mexico. 300 pp.
Sinha, R. N., and W. E. Muir. 1973. Grain Storage: Part of a System. The Avi Publishing
Company, Inc., Westport, Connecticut. 481 pp.
Tropical Products Institute. 1970. Food Storage Manual. Part I: Storage Theory; Part II:
Food and Commodities; Part III: Storage Practice, and Commodity and Technical
Index. Food and Agriculture Organization of the United Nations, World Food
Programme, Rome. 799 pp.


Adams, J. M. 1976. A guide to the objective and reliable estimation of food losses in
small-scale farmer storage. Tropical Stored Products Information 32:5-12.
Adams, J.M., and G.W. Harman, 1977. The Evaluation of Losses in Maize Stored on a
Selection of Small Farms in Zambia with Particular Reference to the Development of
Methodology. Report G-109. Tropical Products Institute, London. 149 pp.
Bourne, M. C. 1977. Post Harvest Food Losses: The Neglected Dimension in Increasing
the World Food Supply. Cornell International Agricultural Mimeograph No. 53.
Cornell University, Ithaca, New York.
De Padua, D. B. 1974. Post-harvest Rice Technology in Indonesia, Malaysia, the Philip-
pines, Thailand: A State of the Art Survey. International Development Research
Centre, Ottawa, Canada.
Frantz, S. C. 1972. Behavioral Ecology of the Lesser Bandicoot Rat, Bandicota ben-
galensis (Gray), in Calcutta. Ph.D. Dissertation, Johns Hopkins School of Hygiene
and Public Health, Baltimore, Maryland.
Guggenheim, H., with Hamidy Hama Diallo. 1977. Grain Storage in the Fifth Region of
Mali: Problems and Solutions. Report to the United States Agency for International
Development. Wunderman Foundation, New York.
Harris, K. L. 1972. Methodology on Assessment of Storage Losses of Food Grains in
Developing Countries. Report prepared for the staff of the International Bank for
Reconstruction and Development, Washington, D.C.
Harris, K. L., and C. Lindblad. 1978. Post-Harvest Grain Loss Assessment Methods.
American Association of Cereal Chemists, St. Paul, Minnesota.
Hopf, H. S.; G. E. J. Morley; and J. R. O. Humphries. 1976. Rodent Damage to Growing
Crops and to Farm and Village Storage in Tropical and Sub-Tropical Regions. Centre
for Overseas Pest Research and Tropical Products Institute, London.
Yashoda, L. U.; M. K. Krishnakumari; and S. K. Majumder. 1977. Specific relationships
among commensal rodents. In: Proceedings of the All-India Rodent Seminar, Ah-
medabad, India, September 23-26, 1975, pp. 89-94. Published by Dr. G. C. Chatur-
vedi, Project Officer of the Rodent Control Project, Sarovar, Sidhpur, India.

Chapter 4

Cereal Grains and

Grain Legumes

The knowledge about the nature and extent of postharvest losses, as noted
earlier, is much more extensive for cereal grains and grain legumes than for
other commodities.
There are a number of reasons for this. In most societies, the durable
commodities are (or have been) the most important in terms of quantity
produced. They are traditionally stored, and security or survival has depended
on keen attention to this process. This has been less true in the case of the
nongrain staples. It is easier to protect dormant dried grain from external
attack by insects or rodents than it is to prevent physiological deterioration
or fungal attack of perishables. Perishables are often seasonal crops that
provide a relatively constant supply of a variety of fruits and vegetables
without storage. Many grow with minimum attention; their husbandry is
therefore much less important and demanding than that of durable staples.
The bulk of harvested cereal grains and legumes passes through a fairly
well-defined series of steps-the postharvest system. After harvest, the crops
are threshed or shelled, dried, stored, and finally processed. Each commodity
has its own variations in this process and some have additional steps that
enlarge the system (rice parboiling, for example), but there are enough
similarities in the flow of durables through the system to enable generaliza-
tion about loss problems. The first part of this chapter is concerned with
these problems; the second part will be a discussion of loss factors specific to
the individual commodities.

General Causes of Postharvest Grain Loss

Preharvest Factors
The genetic characteristics of a grain variety greatly influence the post-
harvest losses it is likely to incur. Traditional varieties are generally well
adapted both to their usual environment and to postharvest handling. The

The grain postharvest system

grains that survive storage and are used in subsequent seasons have evolved
characteristics that favor their survival. These may include, for instance, lower
moisture content in the ripe grain, which then dries more readily, and thicker
seed coat for repelling insects and rodents.
Introduction of varieties selected for high yields has resulted in greater
postharvest losses where the new varieties are not as well adapted to the
postharvest conditions as traditional varieties. This problem should be a
consideration both in selecting high-yielding varieties and in providing for
their postharvest treatment.
Damage to the growing crop may affect its postharvest characteristics, as
may crop protection treatment prior to harvest. In particular, insect infesta-
tion of a maturing crop may increase its vulnerability to loss after harvest;
however, residual insecticide may reduce the extent of postharvest insect

Harvesting Factors
The time at which harvesting occurs has an important effect on the
subsequent storage quality of the grain. Typically, the harvest may be begun
before the grains are fully ripe and may extend until mold and insect damage
are prevalent and shattering has occurred. Grain not fully ripened contains a


higher proportion of moisture, and will deteriorate more quickly than mature
grain because the enzyme systems are still active. If the grain remains in the
field after maturity, repeated wetting from rain and dew at night, along with
drying by the hot sun by day, may cause grain to crack (particularly
long-grain paddy) and may increase the likelihood of insect damage (espe-
cially in maize, paddy, and pulses).
Crops standing in the field after maturity become more liable to harvest
losses. Ripened grain is more likely to be shattered onto the ground during
harvesting. Maize loss may result from the loosening of the husk after it is
ripe and subsequent mold infection or insect attack. The probability of insect
infestation in the field is also likely to increase if the crop stands too long, as
is loss to rodents, grain-eating birds, and other vertebrates.

Threshing and Shelling

Traditional methods of threshing to separate grains from the plant, such as
use of animals to trample the sheaves on the threshing floor-or the modern
equivalent using tractor wheels-may result in loss of grain not separated. This
method also allows impurities to become mixed with the grain, which may
cause subsequent storage problems. The use of flails to beat the grain from
the stalk may also damage the grains or kernels and is not always effective.
Threshing and shelling will contribute to losses if carried out in a manner
that results in cracking of grains.
Modern devices for threshing and shelling may be used incorrectly, or for a
crop for which they were not intended, with excessive breakage of grains.

Traditional threshing with ox-drawn "norag," Lebanon (FAO photo by G. Tortoli)



Drying is a particularly vital operation in the chain of food handling, since
moisture may be the most important factor determining whether, and to
what extent, grain will be liable to deterioration during storage.
Drying is used to inhibit germination of seeds and to reduce the moisture
content to a level that prevents the growth of fungi and bacteria; it can also
retard attacks on the grain by insects and mites.
In developing countries, the methods available to farmers for drying crops
are often limited, usually to a combination of sun and air drying, although
supplemental heat is frequently employed. In many cases, seed grain may be
treated separately from food grain and with greater care. Drying is a complex
process requiring considerable skill and effort on the part of the farmer; the
success with which the grain is preserved over shorter or longer periods
depends to a great extent on the care and attention given to the drying and
subsequent storage. Drying is often complicated by the introduction of
high-yield varieties that mature and must be harvested during wet seasons or
by production of a second, irrigated crop ("double-cropping") that must also
be harvested during the rains. In these cases the grain requires artificial

Sun drying rice, central Java (FAO photo by Jack Ling)
Sun drying rice, central Java (FAO photo by Jack Ling)


drying. The increased production of high-yield varieties and their differing
characteristics may also tax the farmer's ability to handle the grain properly
by traditional methods. Consequently, a new drying and storage procedure
must be adopted or the crop must be sold undried. The alternative may be to
forego the new variety.
Overdrying-which can easily occur in arid regions or after excessive
exposure to sun or other heat-can cause breakage, damage to the seed coat,
bleaching, scorching, discoloration, loss of germinative power, and nutritional
changes. Too-rapid drying of crops with high moisture content also causes
damage; for example, bursting (or "case-hardening"), which causes the sur-
face of the grain to dry out rapidly, sealing moisture within the inner layers.
Underdrying or slow drying (a problem in humid regions) results in deteriora-
tion due to fungi and bacteria, and, in extreme cases, leads to total loss.
Solar technology for artificial drying is receiving attention because of its
negligible running costs in comparison with traditional fuels, which are
becoming not only expensive but, as in the case of firewood (increased
consumption of which is causing deforestation in many areas), are adversely
affecting the environment. However, the fundamental problem with solar
devices is that they do not operate effectively when they are most needed-to
dry grain that must be harvested during a wet spell or during the rainy season.
The Peace Corps-VITA manual, Small Farm Grain Storage (Lindblad and
Druben, 1976), contains descriptions and instructions for constructing a
variety of improved grain dryers-a pit oil barrel dryer, an improved maize
drying and storage crib, a simple batch-type rice dryer, and a number of
simple solar dryers.
Clearly, methods of drying must be selected for the particular climatic,
economic, and social circumstances in which they will be used. This is
especially true where existing drying methods have evolved over long periods
of time to meet community and family survival needs. Alternative methods
should not be recommended without awareness of all possible consequences
to the farmers. Problems affecting the selection of drying methods are
discussed in the second half of this chapter, in the context of individual

Storage Losses
The extent to which deterioration and loss occur in storage depends on
physical and production factors, the storage environment, and biological
factors. Physical factors which contribute to storage loss have been discussed
in the previous section.
In addition, physical damage to the crop during harvest may also affect
storage. Undamaged cowpea pods, groundnut shells, and the husks of paddy
grains also afford the crop a noticeable degree of protection from infestation


by most insect species, though the space occupied reduces the volume that
can be stored.

Storage Environment
Storage conditions have much to do with the rate of deterioration. High
temperature and humidity encourage mold formation and provide conditions
for rapid growth of insect populations. Deterioration is minimal in cool, dry
areas, more marked in hot, dry ones, high in cool and damp conditions, and
very high in hot, damp climates. Climatic conditions during and after harvest
affect the ease with which natural drying may be carried out and may dictate
the need for artificial drying. Seasonal and diurnal temperature differences
between stored grains and the surrounding environment can result in moisture
translocation or migration among quantities of bulk or bag-stored grains or in
condensation of moisture on the grain. Concentration of moisture in grain
can lead to conditions favorable to the development of fungi.
Some climates lessen the residual activity of certain pesticides and can
reduce the effective life of storage containers and structures. Certain struc-
tural materials may alter the effectiveness of different formulations of a given
Deterioration is also related to storage method and management. For
example, cob maize stored in open-sided cribs takes up moisture more rapidly
during the rainy season than shelled maize in mud-walled cribs, so that
conditions for rapid insect development are produced earlier in the storage
season. On the other hand, properly designed open-sided cribs will allow
relatively rapid drying of unhusked ears of maize and reduce losses due to
mold. Traditional pest control methods are often effective in keeping down
infestation levels. For example, some farmers storing pulses and larger grains
will admix a smaller seed or sand with the grains to fill the intergranular
spaces. This effectively inhibits the development of bruchid beetles. Other
farmers use a fire under their storage cribs to repel insects, either through the
effect of the smoke or by keeping the grain dry. The admixture or overlay of
ashes derived from burning various woods or dried animal dung is another
method affording protection against insect attack.

Biological Factors
The principal biological agents of deterioration during storage are insects
and mites, fungi, and rodents.
Losses Due to Insects and Mites. Insect pests are a greater problem in
regions where the relative humidity is high, but temperature is the overriding
factor that influences insect multiplication. At temperatures of about 320C,
the rate of multiplication is such that a monthly compound increase of 50


times the present number is theoretically possible. Thus, 50 insects at harvest
could multiply to become more than 312 million after 4 months.
The nutritive requirements of insects are much the same as those of
vertebrates. Crops with the highest nutritive values for man are also those
most susceptible to insect damage. In certain cases, farmers may keep only
small amounts of a nutritious crop such as beans because they believe damage
and loss to be inevitable. Furthermore, insects often select the most valuable
portion of seeds. For example, four important pests of maize attack the
embryo and reject the starchy endosperm, thus removing the most nutritious
part of the grain as well as destroying the power of germination.
Weight loss is of economic as well as nutritive importance and, in the
absence of effective control measures, insect attack on cereal grains and beans
can be so severe as to reduce the commodity to empty husks and dust. Large
numbers of insects can be expected to produce heavy weight losses, and the
resulting contamination by dead and' live insects and their excreta can be
sufficient to make the commodity completely unpalatable and unacceptable
in the market.
Termites in a grain store can weaken the structure, leading to its collapse.
They will also readily attack the grain.
Table 4:1, prepared by entomologists of the Tropical Products Institute,
shows the main insects and mites that attack and damage stored cereal grains
and pulses in developing countries. The U.S. Department of Agriculture has
recently published a useful compendium of information on stored-grain
insects (USDA, 1978).
Control measures, whether or not insecticides are available, depend first on
storage hygiene. Storage containers must be checked and cleaned as carefully
as possible. Old stored grain should be checked and, if necessary, redried and
cleaned to control existing infestation. New dry grain should be kept separate
from old stored grain because of the risk of cross-infestation. Similarly, stores
should be as remote from the field as possible to reduce the risk of infest-
ation. In addition to store pests, it must be assumed that new grain is infested
from the field and control must include a regular system of inspection and
deterrence to maintain storage hygiene and take control measures where
infestation is observed.
Traditional pest-control systems not involving insecticides are adapted to
local circumstances. Use of local herbs, mixing ash with grain, and smoking
are effective and should be encouraged. As previously stressed, every effort
should be made to build on traditional technology and innovations should be
undertaken with understanding of the social and economic implications. This
is particularly important in the case of insecticides that present severe health
hazards or have other environmental, ecological, economic, and social im-
plications (such as overoptimistic expectations that new technologies will
solve all problems and remove the need for traditional efforts).

Storage Pests of Grains and Pulses

er Family Name
optera Dermestidae Trogoderma granarium Everts


Lasioderma serricorne (F.)

Bostrichidae Rhyzopertha dominica (F.)

Trogossitidae Lophocateres pusillus (Klug)
Tenebroides mauritanicus (L.)




Carpophilus dimidiatus (F.)
Cryptolestes ferrugineus (Steph.)
C. pusillus (Schoen.)
Ahasverus advena (Waltl)
Cathartus quadricollis (Guer.)
Oryzaephilus mercator (Fauv.)
0. surinamensis (L.)
Typhaea stercorea (L.)
Alphitobius diaperinus (Panz.)

A. laevigatus (F.)

Gnatocerus cornutus (F.)
G. maxillosus (F.)
Latheticus oryzae Waterh.

Palorus ratzeburgii (Wissm.)

P. subdepressus (Woll.)

Tribolium castaneum (Hbst.)

T. confusum Duv.

Bruchidae Acanthoscelides obtectus (Say)
Callosobruchus chinensis (L.)

C. maculatus (F.)

Zabrotes subfasciatus (Boh.)
Anthribidae Araecerus fasciculatus (Deg.)
Curculionidae Sitophilus granarius (L.)
S. oryzae (L.)
S. zeamais Motsch.
Lepidoptera Gelechiidae Sitotroga cerealella (01.)
Pyralidae Corcyra cephalonica Staint.
*Ephestia cautella (Wlk.)

E. elutella (Hbn.)

E. kuehniella Zell.
Plodia interpunctella (Hbn.)

Acarina Acaridae Acarus siro L.
Tyrophagus putrescentiae




*Major pest species.

ILE 4:1

Cereal grains and
Cereal grains and
Cereal grains and
Cereal grains
Cereal grains and
Cereal grains
Cereal grains
Cereal grains
Cereal grains
Cereal grains
Cereal grains
Cereal grains
Cereal grains
Cereal grains and
Cereal grains and
Cereal grains
Cereal grains
Cereal grains and
Cereal grains and
Cereal grains and
Cereal grains, grain
products and pulses
Cereal grains, grain
products and pulses
Pulses (esp. beans)
Pulses (esp. peas and
Pulses (esp. peas and
Pulses (esp. beans)
Cereal grains
Cereal grains
Cereal grains
Cereal grains
Cereal grains
Cereal grains
Cereal grains and
Cereal grains and
Cereal products
Cereal grains and
Cereal products
Cereal grains, grain
products and pulses


Many insecticides are becoming widely available in developing countries as
their application is encouraged by suppliers and extension services. Some are
more hazardous to humans, and potentially so to the environment, than
others, but all should be used with great care. Some can be used on seed grain
in high concentrations that would be dangerous where used on food grain.
The grain-storage insecticides are of two main types:

Contact poisons such as dusts, dispersible powders, and emulsions. Some
insecticides, such as BHC, are quite stable and have long residual action;
others, like malathion, usually have little residual action and are used where
human consumption of the grain precludes use of longer-acting chemicals.
Some compounds may be mixed with grain at the time of storage, while
others are used for spraying storage containers or bagged grain. The level of
application and its timing in relation to expected human consumption are
major problems for extension services seeking to improve insect control in
rural grain storage.
Fumigants, which are gases, can penetrate bulks of grain and kill insects
and their larvae living within grains. The negative factors are that all fumi-
gants are safe only when used by trained personnel and that they have no
residual action to protect grain from subsequent reinfestation. A first require-
ment is improvement of the methods of application and more careful moni-
toring of insecticide use to achieve maximum control of infestations. Reports
of insect resistance to chemical insecticides are frequently encountered.
Awareness about the drawbacks of chemical use is also increasing and there
has been renewed interest in traditional nonchemical control techniques and
in developing alternative approaches to pest control.

The principal methods of coping with insect infestations involve cultural
control or manipulation of the environment to make it less favorable to the
insect; breeding resistant crop strains; using natural enemies of insects such as
parasites, predators, and disease vectors; sterilizing insects to interfere with
normal reproduction; and using attractants and repellants.
Cultural control and inbred resistance are not new techniques. Traditional
'methods of controlling insects in storage involve mixing sand, limestone, ash,
or herbs with the grain; in addition to forming a barrier against movement of
the insects through the grain, they abrade or absorb the wax coating of the
insect's protective cuticle, causing a loss of body moisture. In many areas
insects (and rodents) are repelled by smoke from small fires, used either
within granaries to decontaminate them between harvests or under granaries
constructed of permeable materials such as woven plant fibers. The fire also
assists grain drying in situ. In other areas, stored grain is inspected frequently
and is redried in the sun if insects are observed. Hermetic storage, with grain
sealed in impervious containers, is highly effective in excluding insects.


However, the system is difficult to maintain for large quantities and is usually
confined to relatively small amounts of seed grain. (See Note 4-1.)
Along with investigations of newer possibilities for nonchemical biological
control, traditional methods of cultural control should receive greater atten-
tion to increase understanding of their underlying biological basis. The
knowledge gained about both approaches can form the basis of more effective
and safer methods. Since the methods are dependent on manipulation of the
ecology to the detriment of the insects, they are highly location-specific,
which increases the need for research and adaptation of techniques to local
circumstances. This will require long-term study, making it unlikely that there
will be alternatives to replace or greatly reduce insecticide use in the near
Losses Due to Fungi. Fungal attack in storage generally occurs when
drying has been inadequate, when large numbers of insects are present,
causing a temperature rise in the grain, or when the stored crop is exposed to
high humidity or actual wetting. Fungal development does not normally take
place when the moisture content of the commodity is below that moisture
content in equilibrium with a relative humidity of 70 percent. In recent years,
attention has been given to the toxic products of certain fungi, such as
aflatoxin and zearalenone, which are metabolites, respectively, of the fungi
Aspergillus flavus and Fusarium moniliforme. Mycelia penetrate the endo-
sperm of grains, removing nutrients. In many cases the embryo is attacked
first and eventually destroyed.
Fungal spoilage is more serious in those regions with a permanent high
relative humidity or where a season of high humidity coincides with the time
when grain is being dried or kept in store. Microorganisms may multiply and
create heat that can increase in unventilated grain to the point of complete
destruction. However, losses due to fungi are reduced as a result of improve-
ments in drying and storage technology and do not need to be treated
Losses Due to Rodents. Rodent damage to stored food can occur in a
number of ways. The animals not only consume the food (damage to maize
grains is characteristic in that the embryo is usually removed first), but also
foul a large amount with their excretions (which may carry microorganisms
pathogenic to man), destroy containers by gnawing holes that result in
leakage and wastage of grain, and paw into and scatter grain while they eat.
This scattered grain, along with that which leaks from gnawed holes, is
subject to contamination and admixture with impurities. Damage to grain
stored in bulk may be much less than to grain stored on the head or in bags
because rodents are unable to burrow into the bulk.
These problems have recently been reviewed by Hopf et al. (1976) in a
report prepared by the U.K. Centre for Overseas Pest Research and the
Tropical Products Institute. This report analyzes extensive information pro-


vided by a number of governments. It concludes that in most countries very
little is known about the extent of the problem, although some countries
with high losses, such as India, have considerable expertise in this area and
allocate large resources to rodent control.
The three main species of rodent are:

Rattus norvegicus, the Norway, common, or brown rat;
Rattus rattus, the roof, ship, or black rat; and
Mus musculus, the house mouse.

Other species, such as the bandicoot rat (Bandicota bengalensis) are impor-
tant pests in particular areas. Locally, other species can assume greater
Control of Rodents. Techniques for rodent control fall into the following
broad categories:

Rodent exclusion efforts in store construction;
Improved sanitation, including removing food and harborage from the
surroundings or reducing it as much as possible;
Poison baiting, including use of the anticoagulants such as chloro-
phacinone, warfarin, coumarin, diphacinone, and coumatetralyl, and acute
poisons such as zinc phosphide, barium carbonate, red squill, and vacor;
Fumigation with phosphine or other gas;
Trapping and hunting;
Use of cats and dogs; and
Rodent repellents.

Estimates of the effectiveness of these techniques are mixed, sometimes
even contradictory within the same country. Results depend on the thorough-
ness with which the control technique is applied and the length of time it
operates. Control usually is more effective when a combination of methods is
used, particularly those that prevent access to food. The persistence of the
rodent problem is obvious in a report from Israel, where it is fully recognized
and control is vigorous and well organized but where the estimated loss to
crops in the field remains at 5 percent. The problem must be approached with
the recognition that store rodents cannot be controlled unless field rodents
are also controlled.
Observations from the People's Republic of China indicate that well-
organized rodent exclusion, together with sanitation and field control, may
have been rather successful; no published figures are available. Traditional and
modern granaries are reported to be protected by detailed attention to
cleanliness, by physically isolating the granary, by laying concrete on the


Hygienic rice store, People's Republic of China (Courtesy E. S. Ayensu)

area around and underneath the granary (which introduces aspects of behav-
ioral control), and by providing rat barriers at points of potential access
(Ayensu, 1977).
New rodent-control technologies, even simple ones, may meet considerable
resistance at the farm and village level. For instance, local acceptance of
baffles fitted to traditional storage containers has been slow, at best. In this
and similar cases, more research may be needed to determine whether such
unpopular solutions to problems are the most effective. There are reports that
rodents are becoming resistant to rodenticides, although there is little evi-
dence from tropical regions and research on this aspect of control is also
Although many countries fully recognize the seriousness of food loss
caused by rodents-including India, China, and Israel, as we have seen-the
editors of the report of the Centre for Overseas Pest Research conclude that
"the one single fact which emerges most clearly from the survey is the
widespread ignorance of the magnitude of the rodent problem, and of means
to control it."
We have discussed the physical, environmental, and biological causes of
grain losses during storage. Discussion follows of methods of reducing storage
losses, with a brief description of storage practices.


Reduction of Storage Losses
In developing countries, storage on the farm is an important part of the
traditional farming system (the subsistence or nonmarket sector and the
semisubsistence or farm-to-village market sector). It is essential both for
conserving seed for the next planting and for stockpiling staples to feed the
farmer, his family, and his livestock until the next harvest.
Sound storage practice has three elements:

Proper preparation of the grain for storage, including drying and, where
possible, separating out any infested or spoiled grain and other impurities;
Sound storage structures that provide protection from moisture (rain
and ground moisture) or excessive drying and a barrier against insect and
rodent pests and theft; and
An appropriate system of monitoring the quality of the stored grain and
treating and handling it while it is in the store.

Traditional Storage Practices
Traditional systems have evolved over long periods of time to satisfy
storage requirements within the limits of the local culture. Grain for seed is
frequently sealed in gourds or clay containers and kept in the house. Larger
amounts of grain for human and animal consumption are stored in containers
constructed of plant material, mud, or stones, often raised off the ground on
platforms and protected from the weather by roofing material. The design
and materials vary according to local resources and custom.
However, with the exception of sealed containers (including underground
pit stores in drier areas that control insects by limiting the supply of oxygen),
the traditional structures provide only limited protection against insect and
rodent damage, particularly in areas where the climate is warm and humid or
where grain is stored for extended periods.
These traditional grain storage systems have evolved slowly by natural
selection and provide reasonable storage security for the traditional farmer.
This does not mean that losses are necessarily low; it does mean that the risk
of large-scale losses is minimized under traditional decentralized storage
Subsistence or traditional farming systems are being improved by the
introduction of high-yielding varieties of grain, which farmers are encouraged
to grow. However, as a consequence of increased production the traditional
storage system is proving inadequate not only in capacity, but also in
protecting grain from damage, since the new varieties may be more suscep-
tible to insect attack.


There are three approaches to solving traditional storage system problems:

Improving small-scale on-farm storage;
Centralizing grain storage with efficient collection, drying, and large-
scale stores; and
Breeding new varieties that are less susceptible to loss in storage.

Of these approaches, the last two are important long-term possibilities with
political, social, and economic implications that are largely outside the scope
of this report. They will require expanded research efforts, particularly on
socioeconomic aspects of centralizing storage.

New On-Farm Storage Practices
In recent years, this aspect of postharvest technology has been receiving
considerable attention. In East Africa, adaptations have been made to the
traditional design, for example, by fitting "rodent baffles" (Kenya and
Malawi) or mud-plastering cribs for the storage of shelled grain (Zambia and
In Guatemala, India, and Swaziland, prefabricated corrugated or plain
(flat) metal tanks have been in use for storage for a number of years. These
tanks permit fumigation of the grain with hydrogen phosphide tablets, reduce
the probability of reinfestation by insects and rodents, and reduce the rate of
uptake of moisture. In Swaziland at least 40 percent of farmers were using
them by 1976.
Introduction of improved grain bins has not met with the same success in
Ghana and Zambia, where concrete stores proved unacceptable to farmers
because of rising costs, the shortage of materials, and difficulties in construc-
tion. A more recent approach, adopted in Zambia, is to produce a cheap,
easy-to-construct container using readily available materials. The container,
known as the "ferrumbu," incorporates the features necessary for safe grain
storage and should be affordable by emergent commercial farmers.
In Southeast Asia, metal storage containers have been introduced on a
fairly wide scale. Problems have been encountered, however, with drying rice
adequately before storage, providing adequate ventilation, and preventing
stored rice from taking up moisture from the humid atmosphere. These
problems were assigned high priority in reducing losses of stored rice in
Southeast Asia.
Small-scale on-farm grain storage technologies have been compiled by
Lindblad and Druben (1976) in a useful compendium that has been made
widely available to developing countries through the U.S. Peace Corps and
Volunteers in Technical Assistance (VITA).


.: Traditional "gottera," Ethiopia (particularly well made with a good roof cover) (Crown
copyright, Courtesy TPI)

Muddied (left) and unmuddied (right) traditional nkhokwe, Malawi. Note the mud rat
guards on wood framework. (Crown copyright, Courtesy TPI)



Improved dry-store hut, Sehou6 young farmers' club, Bdnin (FAO photo by Banoun/

1-2 tonne metal grain tank, Swaziland, The tank is raised above ground to prevent
bottom corrosion and shaded to minimize moisture migration. (Crown copyright,
Courtesy TPI)


The Lindblad-Druben manual includes discussion of the advantages, dis-
advantages, and construction of various grain-storage methods, including:

Traditional basket storage;
Bagged storage;
Airtight storage-underground pits, including "Thailo" (ferrocement-
lined traditional Thai grain silo), plastic sack storage, metal drums and bins,
and sheet metal silos;
Earthen structures-mud brick silos; and
Cement and concrete structures-cement stave silos and village con-
crete silos.

The selection of storage methods depends on the commodity, the climate,
and the social and economic characteristics of the particular situation. While
improved storage is a prerequisite for sustaining production increases without
increased postharvest losses, the improvements must be carefully attuned to
economic, social, and cultural realities.

Losses During Processing
Primary processing losses occur during:

Threshing and milling;
Parboiling; and
Further processing (baking, brewing, canning, packaging), in which the
losses are important but are not central to the focus of this study.

There is a tendency for processing losses to increase as larger amounts of a
crop are produced and strain the capacity of the traditional processing system.
Maize normally shelled by hand, for example, may be placed in sacks and
pounded with a stick to detach the grains from the cob. Mechanical process-
ing is generally less efficient than manual processing, both because it is
incomplete and because of damage to grains due to their variation in size or
poor adjustment of the machinery. The manual processing efficiency may be
used as the standard against which the efficiency of machinery is measured.
Processing losses are generally specific to particular crops and will be dealt
with under each commodity. There are, however, some general loss problems
resulting from processing. Attitudes towards broken grains vary from society
to society; acceptability of off-color grain due to poor parboiling or drying
varies. In many cases, this simply means that the poorer members of society
have the broken grains and dust, or otherwise lower-quality grain, and there is
little loss. In Pakistan, the Council on Scientific and Industrial Research has
experimented with reconstituting "whole" rice grains from broken grain and
rice powder with good acceptance.


In many societies central milling facilities process grain brought in by farmers
for a price determined by the initial unmilled volume or weight, and there is
thus little incentive for mill operators to reduce subsequent losses due to poor-
ly adjusted equipment or leakage and spillage. Payment may also be in kind,
with part of the milled product or the milling by-products going to the miller.
Where the society is affluent enough that grain adversely affected by
processing is not consumed, the loss is comparatively unimportant to the
central focus of this study.

Individual Crop Loss Problems

Among the hundreds of food crops grown around the world, some two
dozen of them account for approximately 90 percent of all the food produc-
ed. Table 1:1 shows the reported production of major food crops worldwide,
and in the developing countries (according to FAO figures and definitions).
The cereals and legumes, the focus of this chapter, account for more than
half the world's production of food crops and have received most of the
attention in worldwide attempts to reduce postharvest losses. In the main,
developing-country cereals are rice, maize, wheat, sorghum, millet, and bar-
ley. In this section they will be discussed as separate commodities in terms of
the food losses connected with them.


Loss Estimates
Losses of rice in the postharvest food system in developing countries
(Table 4:2) are probably as well characterized as losses for any crop. Yet,
though there are reports of much serious work, there is still uncertainty as to
the magnitude of postharvest losses of rice because of variation in situations
being assessed and because of differences in methodology and definition.
Yet these estimates may be approaching the level of accuracy with which
it is possible to assess losses on a large, nonexperimental scale. There is also a
fair level of consistency from country to country in rice loss figures from simi-
lar situations. These are summarized for Southeast Asia by De Padua (1974):

Harvesting 1- 3 percent
Handling 2- 7 percent
Threshing 2- 6 percent
Drying 1- 5 percent
Storage 2- 6 percent
Milling 2-10 percent


These figures, widely quoted, give a loss range of from 10 percent to over 30
percent (see Note 3-1).
Limited observations on rice in West Africa do not yet specify detailed loss
estimates broken down by stage; however, the overall figures are consistent
with those from Asia (FAO-ECA, 1976). Table 4:2 shows rice weight losses
by percentages for a number of developing countries or areas.

TABLE 4:2 Reported Losses of Rice within the Postharvest System (Based on FAO,
1977b, Figures Unless Otherwise Indicated)
Region Total National
and Percent Production
Country Weight Loss ('000 Tonnes) Remarks
West Africa 6-24 Drying 1-2; on-farm storage 2-10;
parboiling 1-2; milling 2-10 (van
Ruiten, 1977)
Sierra Leone 10 580
Uganda 11 15
Rwanda 9 5
Sudan 17 7 Central storage
Egypt 2.5 2,300 (Kamel, 1977)
Bangladesh 7 18,500
India 6 70,500 Unspecified storage
3- 5.5 Improved traditional storage (Boxall
and Greeley, 1978)
Indonesia 6-17 22,950 Drying 2; storage 2-5
Malaysia 17-25 1,900 Central store 6; threshing 5-13;
c.13 Drying 2; on-farm store 5; handling
6 (Yunus, 1977)
Nepal 4-22 2,404 On-farm 3-4; on-farm store 15;
central store 1-3
Pakistan 7 3,942 Unspecified storage 5
2- 6 Unspecified storage 2 (Qayyum,
5-10 Unspecified storage 5-10 (Greaves,
Philippines 9-34 6,439 Drying 1-5; unspecified store 2-6;
threshing 2-6
up to 30 Malaysia workshop (FAO, 1977c)
3-10 Handling (Toquero et al., 1977)
Sri Lanka 13-40 1,253 Drying, 1-5; central store 6.5;
threshing 2-6
6-18 Drying 1-3; on-farm store 2-6; milling
2-6; parboiling 1-3 (Ramalingam,
Thailand 8-14 14,400 On-farm store 1.5-3.5; central store
12-25 On-farm store 2-15; handling 10
(Dhamcheree, 1977)
Belize 20-30 2 On-farm storage (Cal, 1977)
Bolivia 16 113 On-farm 2; drying 5; unspecified
store 7
Brazil 1-30 9,560 Unspecified store 1-30
Republic 6.5 On-farm store 3; central store 0.3


Rice is relatively difficult to process by hand or with simple manual
equipment, and processing, unlike that for most grains in developing coun-
tries, is widely organized on a collective or centralized basis.
DePadua indicates that because of the complexity of the system, reduction
of losses will require a combination of increased efficiency at each step in the
postharvest system and improved drying, threshing, and milling technology.
Rice postharvest technology is described in detail in the widely available
publication of that title (Araullo et al., 1976) published by the International
Development Research Centre (IDRC). Details of the stages of processing are
described insofar as they relate directly to postharvest losses and loss re-


The bulk of rice produced in Asia and Africa is still grown on small farms
and harvested by hand. Combine harvesters of the type used in Europe and
the United States are unsuited to small farms, and even the smaller Japanese
models have not gained wide acceptance due to high cost, exacting field
requirements, and the high field losses which they cause.
Some major losses are connected with the timing of the rice harvest, that
is, with the maturity of the crop and the effects of postponing the harvest
under certain dry or wet conditions. Delay in harvesting a mature rice crop
leads to lower yields because of lodging and shattering and the exposure of
the ripe grain in the field to birds and rodents. It also leads to postharvest
losses by lowering milling yields and recovery of head grains.
Introduction of new quick-maturing varieties, in major part due to the
breeding and management research carried out at the International Rice
Research Institute in the Philippines, has permitted double-cropping (two
crops per year).
The summer or dry-season rice crop does not have to be dried quickly; if it
is left in the windows after reaping in good weather, there is little damage to
the grain. However, if semidried grain is remoistened by rainfall or a heavy
dew, it may crack and high milling losses will result.
The rainy-season crop creates much more difficulty. To avoid significant
loss, the wet harvest must be threshed, cleaned, and dried within 24 hours
unless predrying facilities are available. It cannot be left in windows or
stacked in the field, since this results in mold, fermentation, and even
germination. De Padua reports that some farmers in the Philippines are
becoming reluctant to plant wet-season crops because of the threshing and
drying problems (TPI, 1978).



Traditional threshing techniques are a frequent cause of loss. They in-

Beating the straws against slats through which the grain falls into tubs or
buckets; and
Threshing by trampling with feet-human or animal-and occasionally
by using tractor wheels or (in Thailand) a tractor-drawn roller.

The Japanese drum thresher-an adaptation of the paddle wire loop thresh-
er-is widely used and has been the first step toward mechanization in India,
Bangladesh, and Burma.
The dry-season crop is frequently dried in the field in windows or stacks
(such as the large rice stack, the "mandola," used in the Philippines) and can
readily be mechanically threshed when relatively dry. A variety of mechanical
threshers and cleaners is found in developing countries, especially in Asia,
frequently based on adaptations of U.S. or Japanese combines. '

Rice-threshing, Indonesia (Courtesy A. A. C. Huysmans)


The wet-season crop tends to choke many conventional mechanical thresh-
ers. Further, the wet conditions make it difficult to bring threshing machines
to the field. Development of satisfactory mechanical wet-season crop thresh-
ing equipment is a priority IRRI research activity; a number of double-drum
threshers have been shown to work satisfactorily and are being introduced
commercially. In many areas the capital cost of these threshers, and the
training and organization their use entails, limits their widespread use, al-
though Samson and Duff (1973) have demonstrated the cost effectiveness of
mechanical threshing, drying, and milling over traditional methods in the
Philippines. Wet-season rice losses due to delayed threshing are likely to be
reduced as mobile mechanical threshers become more widely available.

Rice grain at time of harvest varies in moisture content from low levels to
over 30 percent, depending on the season. This must be reduced to 13-14
percent if the rice is to be stored for any length of time. Inability to do this
quickly leads to rotting of the grain and reduction in milling quality because
of high breakage. It also causes discoloration and loss in quality due to fer-
mentation and heat damage. Mold attack, including possible aflatoxin pro-
duction, may also lead to loss in quantity and quality.
Traditional sun drying of spread-out grain is an effective means of reducing
moisture in the dry season. Wet-season crops require forced-air drying with
heated or ambient temperature air. Commercial driers are designed either to
dry grain in batches, in deep or shallow beds in which the grain may be
stationary or mechanically circulated, or continuously in stages as the grain
flows through the drier. The physical design and operating characteristics,

Field-drying paddy, Indonesia (Courtesy A. A. C. Huysmans)


Winnowing paddy, Hmawbi, Burma (FAO photo)

including such factors as air volume to grain volume ratio and air tempera-
ture, influence the rate of drying and quality of the dried grain.
The mechanical characteristics of driers vary for different kinds of grain,
and great care must be exercised in using a drier for a grain other than the one
for which it was designed. This is particularly important in the case of rice,
which is much more sensitive to thermal stress than other grains and is eaten
as whole kernels. High temperature and low-volume air movement, while
rapidly removing moisture, result in fissuring and a high proportion of broken
grains in milling. The same rate of drying may be achieved at lower tempera-
tures with greater air flow-the important characteristic being the vapor
pressure differential between the drying air and the grain-at less risk of
damage to the grain.
Other characteristics of drying equipment to be considered include the
time required to dry grain from high moisture content and the peak volumes
that must be handled during the harvest season. The strength of the construc-
tion material is important, as paddy rice is highly abrasive and rapidly wears
through sheet metal. Some parts of rice-conveying systems in the People's
Republic of China are frequently constructed of glass, which both counters
the wear and permits visual inspection of the rice flow.


In spite of the evident need for them, driers have not received wide
acceptance in Asia and Africa. Factors mitigating against them include:

The high capital cost of both imported and locally constructed driers
since the limited market has not led to the large-scale manufacture that would
reduce unit cost;
Unsatisfactory performance of some models and lack of experienced
operators, which has resulted in poor milling of dried grain;
Disparity between drying capacity and threshing, milling, and transpor-
tation capacities (with large-capacity central driers only a partial solution,
since they may not be adaptable for many different varieties and grades
received); and
Delays in harvesting, threshing, and transportation that reduce the
benefit of mechanical driers, since drying spoiled grain does not pay for the

For these reasons, it is agreed that the most immediate equipment need is
for low-cost batch-type farm driers fabricated locally from locally available
materials. This simple equipment, properly used, gives dried grain of good
milling quality with a minimum of delay, particularly where the drier can be
moved from the storage or milling plant to the farm to reduce the time
between harvest and drying.
In certain situations, these small batch driers can be used to complement
large central drying plants, particularly at the peak of the harvest season when
the grain has very high levels of moisture. They can be used to dry the grain
partially to about 18 percent moisture level, at which level the grain can be
kept for several days without significant bio-deterioration before being cen-
trally dried to the 13-14 percent moisture level required for storage and
The level of sophistication of the farm-level batch driers can vary accord-
ing to need, capital resources, and construction and operating experience
available. Very simple driers can be built that are also designed to burn rice
husks, thereby reducing operating expenses.
The necessary credit for the farmers to purchase the driers and the in-
frastructure to support their use and maintenance must be available. This
may include pricing policies which encourage production of better quality
grain; the profit incentive will stimulate adoption of the improved tech-
Larger driers can be purchased commercially or designed and constructed
with expert advice from engineers of national agencies or international
organizations such as FAO, IRRI, or the West African Rice Development
Association (WARDA).
Research on the use of rot retardants and natural ventilation drying may
yield useful alternative or supplementary practices (Note 4-2).



Because rice is mainly consumed as intact grains, rice milling, in contrast
to processing of other cereals, is a complicated process; a large number of
operations are required to produce white polished rice grain from harvest
paddy. (Rice is widely called "paddy" prior to milling, and "rice" after
milling, a convention followed here to distinguish between the two forms.)
These operations are parboiling, precleaning, hulling, husk separation, paddy
separation, and whitening and polishing.
At each stage, losses are incurred due to the inherent efficiency limits of
the process and, of course, due also to inefficient operation. These losses-
and the processes themselves-are reported to be priority areas for atten-
tion and improvement and are described in Note 4-3.


Much of the traditional rice-growing area of the world lies in the humid
tropics, with a climate characterized by high average temperatures (around
300C) and relative humidity (around 85 percent).
Under these conditions, grain (whether stored as paddy or as milled rice) at
equilibrium tends to absorb moisture from the atmosphere at the more humid
times of day. Unless the storage is adequately ventilated, spoilage may result
from increased respiration of molds in the grain, accompanied by local
heating and possibly fermentation.

*7 ,.' ,4 V.
.-- '5

Traditional rice-storage houses, Laos (FAO photo by H. A. Wirtz)
Traditional rice-storage houses, Laos (FAO photo by H. A. Wirtz)


-' ^ '* La A za

Improved traditional paddy storage, India (Courtesy A. A. C. Huysmans)

In closed structures, there can be a problem with condensation of mois-
ture, which may be given up by moist grain at the hottest part of the day and
condense on the walls during the night, causing local wetting of the grain.
Aeration (forced ventilation) permits equalization of moisture content
throughout the stored grain and also reduces the temperature of the grain by
evaporation. The drop in temperature also tends to reduce respiration and
Paddy is commonly stored either in the form of bundles of panicles, in
sacks or plastic bags, or in bulk storage. The sacks or bags provide a means of
separating varieties for specific milling requirements. However, they deter-
iorate with use and allow access to insects and rodents, particularly if not
properly stacked, if handled with hooks, and if the store hygiene is not
adequate. Bulk storage, if properly organized, is efficient and relatively
inexpensive. However, its efficient operation requires considerable capital
investment and trained manpower, both of which may not be available.
Storing rice as paddy has advantages over storing milled grain, particularly
where storage facilities are less than completely adequate. The protection
afforded the kernel by the husk against insect, fungal, and even rodent attack,
as well as the problems of storing poorly milled rice, accounts for the fact
that the bulk of harvested dried paddy is stored in this form before milling,
although this depends to some extent on the local economic situation and
supply and demand for paddy and milled rice at different times in the season.


-L -
"Kanduch" lath and clay rice store, Chasemabad District, Gilan Province, Iran. Note the
log-disc rodent baffles. (FAO photo)

Washing Losses

Rice is commonly washed prior to cooking, and in Korea, Cheigh et al.
(1977) estimate weight losses due to washing local varieties of polished rice at
over 2 percent. While it can be argued that this is not, strictly speaking, a
postharvest loss as such losses are defined in this report, since the food is in
the hands of the final consumer and loss takes place immediately prior to
consumption, it has postharvest loss implications. The loss of nutrients
accompanying the total solids loss is similarly not a postharvest weight loss,
but is clearly important. The authors suggest that improved postharvest
handling that delivers sanitary rice to the household and makes washing
unnecessary could minimize these substantial losses.


Economics of Improved Processing
Studies have been made at IRRI on the magnitude and nature of field
grain losses in paddy production. Samson and Duff (1973) indicate different
levels of loss between varieties, wet and dry seasons, and varying moisture
contents at time of harvest. A related series of trials investigated losses
ascribable to handling, bundling, stacking, and field drying. Total paddy
losses are given as approximately 1-3 percent for the harvest operation and
2-7 percent for the intermediate handling steps.
A comprehensive survey of rice milling and field level operations was
undertaken in three regions of the Philippines, including 180 rice mills and
approximately 600 farms. Based on the results, a series of pilot trials was
undertaken to assess the impact of different systems of technology at the
farm and rice mill level.
These trials compared the traditional methods commonly employed by
farmers (threshing manually and sun-drying prior to storage and milling) with
various systems of improved technology: mechanical threshing, batch-drying,
and combinations of each with manual threshing and sun-drying.
Losses from manual threshing were found to be three times greater than
losses from mechanical threshing. Increases in yield of about 12 percent were
observed using either the drier or thresher in combination with traditional
methods, and using both in sequence gave the greatest increase in output. A
large part of this increase is ascribed to the reduction in time between harvest
and threshing.
The improved technology also increased the quality of the grain, largely by
reduction of broken, fermented, discolored, and immature grains due to the
improved drying and reduction in the time between harvest and drying. Total
expenditure per tonne was estimated to be twice as high for traditional
methods as for the improved methods evidently much favored by farmers.


Loss Estimates
FAO-published figures (1977b) report average maize losses from 9.6 per-
cent to 20.2 percent, mainly in storage (either unspecified or on-farm as
opposed to central storage) and due primarily to insect damage, followed by
fungus and rodent damage. However, the data is markedly inadequate, and as
the FAO report concludes, "the estimates of losses of durable commodities
and the methods by which they are derived were inadequately refined." Much
painstaking work that has been published on farm-level maize storage is of
limited use in determining weight losses because of the difficulty of measur-
ing and interpreting losses due to "insect damage" reported as a percentage of


damaged grains. Adams (1977) notes the lack of information from Central
and South America in contrast to the considerable attention paid to cereal
losses in most regions of Africa.
Maize presents particular problems of loss estimation because it can be
stored either on the cob or shelled, affecting the subjective evaluation of what
is edible. Storage on the cob enables a process of selection at the time of
shelling; individually damaged grains may be separated as the grain is shelled,
or the cob may be considered so heavily damaged that it is rejected, good
kernels and all. The correlation between visible insect damage and weight loss
varies according to the type and length of infestation. Accurate estimates of
losses are impossible unless clear definition of these factors accompanies the
numerical data.
The preferred methodology for determining maize losses involves weighing
standard volumes of shelled grain. Losses would be measured on a dry-weight
basis and reported on a typical moisture content that is acceptable in the
marketplace. This moisture content should ideally become a "standard" that
would be widely adopted.
Table 4:3 gives an indicative compilation of reported estimates in different

Harvesting Factors
Postharvest losses are influenced by harvesting practice and field infesta-
tion by pests.
In many areas, standard practice includes turning down the ripe ears on
the stalk, where they are left to dry further before collection; in traditional
varieties, the corn husk completely encloses the grain and protects it from
insect attack. The husk also protects the ear from exposure to moisture after
it is turned down.
Resistance to field infestation by insects is low in many high-yielding
varieties. Increased grain production on the cob generally results in both more
and larger grains that may be incompletely covered by the corn husk and are
more susceptible to field attack by insects and birds. The effects of these
attacks may be carried over to the storage phase.

Damage in shelling is proportional to the moisture content of the grain.
Maize shelling is traditionally accomplished by hand and this method, though
hard, tedious, labor-intensive work, is efficient in stripping the cobs and in min-
imizing damage to the grains; it also permits hand separation of damaged or in-
fested grain from sound grain. Increased production increases the amounts of
grain to be shelled and this can strain the capacity to shell the dried cobs by


TABLE 4:3 Reported Losses of Maize within the Postharvest System (Based on FAO,
1977b, Figures Unless Otherwise Indicated)

Region Total National
and Percent Production
Country Weight Loss ('000 Tonnes) Remarks

Benin 8-9 221 Traditional on-farm storage; 6
months improved silo storage
(Harris and Lindblad, 1978)
Botswana 62 Insect damage
Ghana 7-14 395 (Rawnsley, 1969)
15 8 months storage (Hall, 1970)
Ivory Coast 5-10 120 12 months stored on cob (Hall,
1970; Vandevenne, 1978)
Kenya 10-23 1,360 4-6 months central storage
12 Hybrid maize, hotter regions, 6
months (De Lima, 1973)
Malawi 6-14 1,200 Drying 6; on-farm store 8 (TPI, 1977;
Schulten, 1975)
min. 10 Hybrid
Nigeria 1-5 1,050 On-farm storage
5.5-70 6 months on-farm storage (FAO:
ECA, 1977)
Rwanda 10-20 60 On-farm storage
Tanzania 20-100 1,619 Unspecified storage
9, 14, 67 3, 6, and 9 months (Mushi, 1978).
Togo 5-10 135 6 months central storage (Tyagi
and Girish, 1975)
Uganda 4-17 623
Zambia 9-21 750 On-farm storage (Adams and Harman,
India 6.5-7.5 6,500 Central storage, 7.5 (Agrawal, 1977)
Indonesia 4 2,532
Pakistan 2-7 70
Belize 20-30 20 Traditional on-farm storage (Cal,
Brazil 15-40 17,929 Farm storage
Republic 19 49 Farm storage, 15; processing, 1
Honduras 20-50 289 Traditional storage, poor facilities
(Balint, 1977)
Mexico 10-25 8,945
Nicaragua 15-30 201
Paraguay 25 290 (Martino, 1977)
Venezuela 10-25 532

hand. Methods of shelling quantities of cobs include beating bagged cobs with
a stick, which results in increased loss due to incomplete stripping of the cobs
and damage to the grain. Mechanical shelling losses are relatively low when
the equipment is adjusted and operated competently and grain moisture levels
are low. Wooden hand-held maize sellers developed by the Tropical Products
Institute (Pinson, 1977) offer an efficient intermediate-level technology to


Corn-husking, N'Kolbisson, Cameroon (FAO photo)

increase manual shelling capacity. The dry maize cob is held in one hand, the
sheller in the other, as illustrated. As the cob is pushed into the sheller, the
ridges pull out the grain.

Drying and Storage

Harvested ears of corn are normally stored and dried at the same time and
these two functions should be considered together.
In developing countries there are a number of on-farm traditional methods
for drying and storage of maize.


Wooden, hand-held, maize sheller (Crown copyright, Courtesy TPI)


Ewe barn, Ghana (Crown copyright, Courtesy TPI)


In Ghana, for instance, these include the Ewe barn, a raised circular
platform on which ears are piled up to form a cylinder and roofed with
thatch, and the Ashanti crib, a raised rectangular structure of wood or
bamboo that is also thatched. Losses are estimated at 7-14 percent over 3-6
months (Rawnsley, 1969).
In a number of places in West Africa and elsewhere, ears are hung to dry
from horizontal poles protected from the rain or from branches of trees.
In Kenya and Tanzania, stores for shelled grain are constructed from
woven branches and the basket structure supported by a strong platform
raised on poles or stones. In humid areas the maize is stored in a special loft
or crib over the cooking fire, which deters insects and reduces humidity.
In a number of countries storage baskets may be plastered with mud.
Sometimes they are completely sealed to exclude infestation, in which case it
is usual to open the store periodically and redry the grain in the sun.
The use of hessian sacks for storing shelled grain is widely reported. A
limitation is that the sacks are vulnerable to insect, mold, and rodent damage

^ "' "

Traditional mud-walled stores of the low savannah zones, West Africa (Courtesy W.


when left on the ground in the corer of a hut for extended periods, although
this may be prevented by frequent inspection.
Other containers commonly used to store small quantities include gourds,
pots, tins, and small baskets.
In general, the traditional methods of storage work well as long as they are
in balance with the rest of the farming system. The inherent quality of local
maize varieties (including hardness of the endosperm and low moisture
content), along with storage on the cob with husk intact, help to protect the
grain from insect attack. A long process of natural selection has led to the
survival of varieties that provide a reasonable return to farmers for their
farming and storing efforts. This can be seriously disrupted by introduced
changes; in addition to new crop varieties with their different grain prop-
erties, the amount of grain to be stored generally increases and may place a
strain on the traditional storage facilities or on the time available to the
farmer and his family for processing it.
Improvements to Traditional Drying. Drying is often a problem, particu-
larly when, as in West Africa, the crop matures during a period of high
humidity and the effectiveness of natural drying of large quantities of maize
is uncertain.
The African Rural Storage Center at Ibadan, Nigeria, recommends the
construction from locally available materials of improved drying cribs de-
signed to make best use of free ventilation drying effects. (Note 4-2.) About
4 months of crib drying of dehusked maize, to which insecticides are applied
externally, is the technique recommended over field drying or the storage of
shelled grain immediately after harvest. Crib drying reduces losses due to both
fungi and insects to about 2 percent after 4 months' storage. There is no
evidence to show that small-scale mechanical driers are a better answer to


Conventional crib with rat guards (Courtesy W. Boshoff)


Crib constructed entirely from local materials (Courtesy W.

Large conventional crib constructed mainly from local materials (Courtesy W. Boshoff)



TPi :ommended mesh-sided crib, Swaziland. Note the corrugated iron roof and metal
rat guards. (Crown copyright, Courtesy TPI)

drying needs, and the fuel supply from firewood or fossil fuels makes this
method environmentally and economically less attractive than free ventilation
methods for farm- and village-level drying.
Nyanteng (1972) notes the problems of organizing cooperatives to carry
out drying and storage and the difficulties of introducing improvements
through grain marketing cooperatives in Ghana. The problems stem partly
from inappropriate technology and partly from poor management.
Reduction in losses due to insects has been reported when properly dried
maize is stored shelled. This controls Sitotroga by restricting its movement
through the bulk of the grain and its opportunity to deposit eggs. In the
shelled grain, damage is confined to the exposed surfaces. Sitophilus is readily
controlled by insecticides suitable for direct application to grain, such as
malathion. Crib drying with insect control, followed by storing shelled grain
in impervious containers, has been reported as a successful combination for
reducing losses in humid regions of West Africa and Zambia.


The production of wheat continues to increase in the main producing
countries-India, Pakistan, and Mexico-and the problems of harvesting,
threshing, transport, marketing, storage, and processing have increased pro-
portionally. However, these have been accompanied by extensive efforts to
improve the postharvest system in these countries, and research, training, and
infrastructure are relatively sophisticated compared to those in other

Loss Estimates
Relatively few loss problems suffered by all grains are specific to wheat.
Loss reports for wheat average 10 percent, with the major causes being insects,
rodents, and mold during storage, particularly where overproduction has
strained storage facilities to their limits. As noted, rodents are a major prob-
lem in India.
Milling losses are not reported to be a serious problem. As with rice, they
result from inefficient operation or maintenance of machinery, often due to
inexperienced operators or difficulty in obtaining spare parts. The roller
milling of wheat grains is a highly complex technical process that must be
carefully adjusted to achieve efficient milling of wheat grain to flour of
various qualities, plus yielding germ and bran. Wheat milling differs from rice
milling in that the product is ground flour rather than intact kernels and grain
breakage is not a major issue. Unlike rice processing, wheat processing is not
reported to be an area where substantial postharvest loss occurs or that
requires priority research in developing countries.
With wheat, the primary need is careful precleaning prior to milling to
remove all damaged, spoiled, or infested grains and dirt and debris. This
ensures high-quality flour and protects the machinery from damage.


Threshing and winnowing are still done in open yards rather than in
combines. The grain is exposed for an extended period; considerable losses
due to birds, rodents, and spillage occur, and losses due to high moisture,
molding, and fermentation may result if it rains. Inefficient operation of
mechanical threshers breaks the grain and makes it more susceptible to insect
attack or threshes it incompletely.

In addition to storage problems common to all harvested grains, wheat
may be rendered unfit for malting or bread making because of spoilage during


storage, particularly due to mold. Indian government reports on storage losses
of wheat (Krishnamurthy, 1972) give figures of approximately 2.5 percent
loss due to rodents and a slightly higher percentage due to insects. Small-scale
farm-storage losses vary, but are typically around 10 percent.

Processing Losses
Use of wheat and wheat products is increasing in developing countries in
response to the demand for bread. However, wheat flour is put to other uses
besides bread. In a number of countries, unleavened bread ("hubus," "cha-
pattis," "puris," etc.) is made from varying grades of flour.
While white flour technically comprises about 63-80 percent of the whole
wheat grain, practical milling considerations in current mills require an
extraction rate of about 75 percent to produce a white flour suitable for
western bread making. More widespread use of higher extraction brown flour
(such as the 95 percent "atta" in India) would reduce milling losses.

Barley is produced in cool, upland areas and is a major crop in Korea,
China, India, Iran, Syria, Turkey, and Ethiopia. Its processing is similar to
that of rice. An important use of barley is for sprouting (malting) and
fermentation to produce beer and other alcoholic beverages, and considerable
quantities are imported by developing countries for this purpose.
Drying of the grain prior to malting must be done carefully to ensure that
germination and malting are not affected.
There are few reports in the literature of losses of barley in developing
countries, and it appears that there is no special problem peculiar to barley.
This may reflect, in part, the fact that it is mainly produced in the cooler
countries, the care with which the brewing industry handles the crop, and the
relatively few stages of processing it requires.

Millets and Sorghums
Millets and sorghums are grown in semiarid regions, and although their
annual reported production, approximately 20 million and 30 million tonnes
respectively (FAO, 1977a), represents only 6 percent of total cereal produc-
tion, they are the main staple in drier regions of Africa, the Middle East, India
and Pakistan, and China.
In developing countries, most millet and sorghum production is still at the
farm and village level, and postharvest technology is unimproved compared
with that for the major cereals.


TABLE 4:4 Reported Losses of Wheat and Barley, Millets, and Sorghum within the
Postharvest System (Based on FAO, 1977b, Figures Unless Otherwise


Commodity Total National
and Percent Production
Country Weight Loss ('000 Tonnes) Remarks

Pakistan 5-10 8,500 On-farm storage 5-10; milling 2;
central store 5 (Qayyum, 1977;
Greaves, 1977; Chughtai, 1977)
12 Unspecified storage
India 8-25 24,000 (Amla, 1977; Agrawal, 1977)
2-52 Farm storage to 45; threshing 1;
central storage 8
Rhodesia 10 2,571 On-farm storage (Howden, 1977)
Sudan 6-19 880
Bolivia 16 833 Store 7; drying 3
Brazil 15-20 906 Storage 1-4
Pakistan 9 130 Unspecified storage 7; processing 2
Bolivia 14 80 Drying 2; unspecified store 6;
transport and distribution 4
Sudan 17 Central store
Mali 2-15 804 On-farm store 2-4; central store
10-14 (Guggenheim, 1977)
Nigeria 0.1-0.2 3,200 On-farm storage
Rhodesia 10-15 564 On-farm storage (Howden, 1977)
Sudan 14 450 Central store
Zambia 10 30 On-farm storage
India 7-10 9,600 Drying 2-5; farm storage 5
(Chaturvedi, 1977; Agrawal, 1977)
Pakistan 7.5 503 Storage 5; processing 2.5
Nigeria 0-37 3,680 On-farm over 26 months (Hall, 1970)
Rhodesia 25 716 On-farm store (Howden, 1977)
Sudan 6-20 1,800 Central storage
Zambia 0-10 46 Local varieties, negligible; high-
yielding varieties 10 (TPI, 1977)
India 7.5 544 Unspecified storage
Indonesia 4.0 -
Pakistan 7.0 621 Storage 5; processing 2

Loss Estimates

Postharvest losses of millets and sorghums have received relatively little
attention. Reported losses have often cited damage rather than weight loss.
Loss estimates have indicated relatively moderate levels occurring during
storage, 1-5 percent (Spencer et al., 1975). Other losses occur during harvest
and as the grain dries in the field.



Because harvesting is carried out under dry conditions, millet and sor-
ghum crops are commonly left standing in the field to dry for a period.
Sorghum stalks may be tied together at the top in threes or fours to prevent
the dry stalks from lodging.
This field-drying period may extend for a considerable time, during which
the grain is exposed to bird, rodent, and insect attack, including termites.
Although in many cases these may be considered as preharvest problems,
depending on local variations in handling, the results affect postharvest
deterioration of stored grain.

Millet and sorghum grain is frequently stored on the head and threshed as
required, except for seed grain, which may be threshed and sealed in small
containers to be kept in the house for special security. In other cases,
however, seed grain is kept unthreshed and hung from the roof.

Millet Storage
Guggenheim (1977) and Spencer et al. (1975) describe traditional and
central warehouse storage of millets in Mali and Senegal and report losses
averaging 2-15 percent. Losses in traditional stores were lower than in the
central warehouses.
The difference between estimates of loss under traditional and central
storage results from a number of factors. In Mali, farmers may be assigned a
quota of grain that they must sell to the government; they are prepaid for

Sun drying millet, southern India (Courtesy A. A. C. Huysmans)


their quota and provided with empty sacks in which to deliver the shelled
grain, but since they are not paid for the transportation, there is no incentive
to speed the delivery of the bagged grain to the collection centers. The
collection centers themselves have insufficient, often ill-designed space; even
specially designed stores may be poorly constructed. The bagged grain may be
left uncovered on open ground or kept in old storerooms or school buildings
or the extension agent's residence, together with his animals. And since the
grain that farmers sell to the government is commonly the oldest available, or
of a quality they do not wish to store themselves, it may not be very durable
at the outset. As a result, comparatively high losses (10-15) percent are
encountered in the government system.
Dogon traditional granaries are built of rock, wood, and banco (clay mixed
with millet chaff) about 2.5-3.33 m high, depending on the terrain, excluding
their conical grass thatch roofs. They have wooden foundations that lift
the mud structure 30 cm from the ground to protect it from moisture.

4 I ,;T77


Dogon granary and pounding millet, Mali (Courtesy H. Guggenheim)


Good management of granaries is fundamental to keeping postharvest
losses low. Millet heads are roughly classified during harvest: very poor and
aborted grain is not cut; poor grain is cut but kept apart; good grain is
prepared for storage. The grain to be stored is spread on the flat terraces of
the houses for drying. The very best is reserved as seed, usually kept under
the kitchen roof, protected by smoke against insect attack. The remainder
is moved to the granary 6 weeks to 3 months following harvest, in Decem-
ber or January, by which time it is very dry-typically below 10 percent

Grain storage basket, northern Ivory Coast (FAO photo by A. Tessore)

University of Florida Home Page
© 2004 - 2010 University of Florida George A. Smathers Libraries.
All rights reserved.

Acceptable Use, Copyright, and Disclaimer Statement
Last updated October 10, 2010 - - mvs