Front Cover
 Title Page
 Directors and staff
 Table of Contents
 Chapter 1: The Puebla area - Its...
 Chapter 2: Project operations and...
 Chapter 3: Agronomic research
 Chapter 4: Maize variety impro...
 Chapter 5: Technical assistance...
 Chapter 6: Organization of the...
 Chapter 7: The agricultural service...
 Chapter 8: Evaluation procedur...
 Chapter 9: Farmer adoption of the...
 Chapter 10: Impact of increased...
 Chapter 11: A benefit - Cost analysis...
 Chapter 12: Training of staff for...
 Chapter 13: Promotion of regional...
 Chapter 14: The Puebla approach...
 A benefit: Cost analysis of the...
 Some publications on the Puebla...

Group Title: Puebla Project : seven years of experience, 1967-1973 : analysis of a program to assist small subsistence farmers to increase crop production in a rainfed area of Mexico
Title: The Puebla Project
Full Citation
Permanent Link: http://ufdc.ufl.edu/UF00053899/00001
 Material Information
Title: The Puebla Project seven years of experience, 1967-1973 : analysis of a program to assist small subsistence farmers to increase crop production in a rainfed area of Mexico
Physical Description: ix, 116, 2 : ill. ; 28 cm.
Language: English
Creator: International Maize and Wheat Improvement Center
Publisher: Centro Internacional de Mejoramiento de Maiz y Trigo
Place of Publication: México
Publication Date: 1974
Subject: Corn -- Mexico -- Puebla (State)   ( lcsh )
Agricultural innovations -- Mexico -- Puebla (State)   ( lcsh )
Genre: bibliography   ( marcgt )
non-fiction   ( marcgt )
Bibliography: Bibliography: p. 117-118
Funding: Electronic resources created as part of a prototype UF Institutional Repository and Faculty Papers project by the University of Florida.
 Record Information
Bibliographic ID: UF00053899
Volume ID: VID00001
Source Institution: University of Florida
Holding Location: University of Florida
Rights Management: All rights reserved by the source institution and holding location.
Resource Identifier: aleph - 000336973
oclc - 03710057
notis - ABW6632
lccn - 77376660

Table of Contents
    Front Cover
        Front Cover 1
        Front Cover 2
    Title Page
        Page i
        Page ii
        Page iii
    Directors and staff
        Page iv
        Page v
    Table of Contents
        Page vi
        Page vii
        Page viii
        Page ix
        Page x
    Chapter 1: The Puebla area - Its land and people
        Page 1
        Page 2
        Page 3
        Page 4
        Page 5
        Page 6
        Page 7
        Page 8
        Page 9
        Page 10
    Chapter 2: Project operations and coordination - An overview
        Page 11
        Page 12
        Page 13
        Page 14
        Page 15
        Page 16
        Page 17
        Page 18
    Chapter 3: Agronomic research
        Page 19
        Page 20
        Page 21
        Page 22
        Page 23
        Page 24
        Page 25
        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
    Chapter 4: Maize variety improvement
        Page 39
        Page 40
        Page 41
        Page 42
        Page 43
        Page 44
    Chapter 5: Technical assistance to farmers
        Page 45
        Page 46
        Page 47
        Page 48
        Page 49
        Page 50
        Page 51
        Page 52
        Page 53
        Page 54
    Chapter 6: Organization of the farmers
        Page 55
        Page 56
        Page 57
        Page 58
        Page 59
        Page 60
        Page 61
        Page 62
    Chapter 7: The agricultural service agencies
        Page 63
        Page 64
        Page 65
        Page 66
        Page 67
        Page 68
        Page 69
        Page 70
        Page 71
        Page 72
    Chapter 8: Evaluation procedures
        Page 73
        Page 74
        Page 75
        Page 76
        Page 77
        Page 78
    Chapter 9: Farmer adoption of the maize recommendations
        Page 79
        Page 80
        Page 81
        Page 82
        Page 83
        Page 84
        Page 85
        Page 86
        Page 87
        Page 88
    Chapter 10: Impact of increased production on income, employment and general welfare
        Page 89
        Page 90
        Page 91
        Page 92
        Page 93
        Page 94
    Chapter 11: A benefit - Cost analysis of the Puebla project
        Page 95
        Page 96
        Page 97
        Page 98
        Page 99
        Page 100
    Chapter 12: Training of staff for regional production programs
        Page 101
        Page 102
    Chapter 13: Promotion of regional production programs in other areas
        Page 103
        Page 104
        Page 105
        Page 106
    Chapter 14: The Puebla approach - A synthesis
        Page 107
        Page 108
        Page 109
        Page 110
        Page 111
        Page 112
        Page 113
    A benefit: Cost analysis of the agronomic research program
        Page 114
        Page 115
        Page 116
    Some publications on the Puebla project
        Page 117
        Page 118
Full Text


Amnlys of a Propm to Asust Small SubdistMc Far.m to
Increase Crop Production in a Rainfed Area of Mexico

Correct citation: Centro Internacional de
Mejoramiento de Ma(z y Trigo. 1974
The Puebla Project: Seven years of expe-
rience :1967-1973. El Batan, M6xico.

L o n d r s 4 0 Aparlado Postal 6-641 M6xico 6, D. F., M6xico



Analysis of a Program to Assist Small Subsistence Farmers
to Increase Crop Production in a Rainfed Area of Mexico

Ministry of Agriculture, Government of Mexico
Government of the State of Puebla

International Maize and Wheat Improvement Center (CIMMYT)
Graduate College, Chapingo, M6xico.
OPERATION (Fertilization Committee of the State of Puebla)
Ministry of Agriculture
Government of the State of Puebla
The national fertilizer company, "Guanos y Fertilizantes"
Ministry of Water Resources
Department of Agrarian Affairs and Colonization
National Agricultural Credit Bank
National Ejidal Credit Bank
Agricultural Bank of the South
League of Agrarian Communities
Association of Small Private Farmers
National Agricultural Insurance Agency
National Seed Producing Agency
National Marketing Agency
Fertilizer distributor, "Impulsora de Puebla"
Fertilizer distributor, "Agr6nomos Unidos"
Fertilizer distributor, "Fertilizantes Olmeca"
Fertilizer distributor, "Fertiton de Puebla"


This report documents the first seven years of the Puebla Project its philosophy,
objectives, organization, operation and accomplishments. It was prepared by staff mem-
bers and advisors, most of whom have been associated with the Project since its begin-
In analysing progress during these initial years, this report reflects the Project's multi-
disciplinary team approach. An introductory section states Project objectives, initial
conceptualization, and the prerequisites for locating the Project area. The characteristics
of the area its land and people are described next, followed by an overview of the
evolvement of Project operations from 1967 to 1973. Chapters 3 and 4 detail the
methodologies, field operations, and accomplishments of the maize research programs;
Chapters 5 and 8 describe the methodologies and field operations of the technical as-
sistance and evaluation programs.
Subsequent discussion (Chapters 6, 7, 9, 10) provides an examination of Project activi-
ties in organizing farmers, improving the operations of service institutions, persuading
farmers to adopt new technology, increasing maize yields, and improving the general
well-being of the farmers. The benefits attributable to the Project are compared with costs
in Chapter 11. The Puebla Project's role in getting similar programs underway in other
parts of Mexico and other countries is outlined in Chapters 12 and 13. In its final chapter
(14), the report seeks to summarize what has been learned about the organization and
operations of successful regional agricultural development projects.
In terms of the Project's long-term goals, the operations at present are seen to be in
mid-process, with contributions already extending far beyond the Puebla Valley. The
initial focus on increased maize production has widened, as originally envisioned, to
encompass the broader development objectives of increased net income, greater
employment opportunities, and improved general welfare of rural families.
During the period covered by this report, from 1967 though 1973, the Project was
administered by the International Maize and Wheat Improvement Center (CIMMYT) and
operated jointly by that institution and members of the Chapingo Graduate College. In
early 1974, responsibility for the Project passed to the Mexican Ministry of Agriculture
where it is presently being operated by the Chapingo Graduate College. Members of the
CIMMYT staff who were functioning as advisors to the Project in 1973 have moved to the
Graduate College where they are continuing their advisory role to the Project and are
participating in academic programs to train people in new approaches to increasing crop
production and improving the quality of life in disadvantage rural areas.


The Project operations and accomplishments have depended upon the firm support of
the highest authorities of the Mexican Government. The Ministry of Agriculture has(given
strong backing at both the federal and state levels. The Government of the State of
Puebla has supported the Project as an effective means for beginning an economic and
social transformation of the area. The cooperation and support of governmental author-
ities and private enterprises has made it possible to improve credit facilities and to make
more readily available the essential inputs and services.
The Chapingo Graduate College, through its Departments of Agricultural Communica-
tions, Statistics, Soils, and Genetics, has provided advice and direct service. The National
Institute of Agricultural Research, through its Maize Department, has provided genetic
materials for developing new high-yielding varieties for the area. CIMMYT has provided
technical advice and direct assistance through its Maize Program and Communications
Department and also has provided an administrative structure with a minimum of the
bureaucratic impediments that might restrict the functioning of the Project.
The Rockefeller Foundation made the initial grant for the Project and contributed a
total of $559,851 (see note below) to its support during the seven-year period,
1967-1973. During this period, CIMMYT contributed $332,737 for the operation of the
Project, and $32,457 was provided by the Graduate College at Chapingoand institutions
in Puebla. The total operating cost of the Project for the seven years was $925,045.
Other activities related to the Project and their costs were: (1) Consulting service for
the Project team -$87,166; (2) Consulting service for similar programs in Mexico and
other countries -$112,390; and (3) Training of staff for other programs -$168,158. These
costs for the operation of the Project and related activities do not include a charge for
indirect costs (overhead).

EDITORIAL NOTE: This report was prepared by Dr. Leobardo Jimenez Sanchez, Dr.
Antonio Turrent Fernandez, Dr. Heliodoro Diaz Cisneros, Lic. Francisco Garcia Hernan-
dez, Dr. Gregorio Martinez Vald6s, and Dr. Reggie J. Laird, with ample assistance from
other members of the Puebla Project staff.
Dr. Alain de Janvry, Dr. John Pesek, Dr. Delbert Myren, and Dr. Ralph Cummings, Jr.
reviewed the manuscripts and contributed significantly to improving their content.
All references to dollars in this report are expressed in terms of United States cur-

Directors and Staff

Honorary Directors
DR. OSCAR BRAUER HERRERA Secretary of Agriculture
SR. GUILLERMO MORALES BLUMENKRON Governor of the State of Puebla

Advisory Board
HECTOR PORRAS HOWARD, Ing. Agr. General Agent for the Federal Ministry of Agri-
culture in the State of Puebla, 1967-1969.
RAFAEL TERRAZAS LOYOLA, Ing. Agr. General Agent for the Federal Ministry of Ag-
riculture in the State of Puebla, 1970-1971.
CARLOS TRILLO GUIDO, Ing. Agr. General Agent for the Federal Ministry of Agricul-
ture in the State of Puebla, 1972-1973.
ALFONSO VILLARREAL, Ing. Agr. Director General for Agriculture of the State of
OSCAR BRAUER HERRERA, Ph. D. Director, Graduate College, Chapingo, Mexico
LAURO BUCIO ALANIS, Ph. D. Director, Graduate College, Chapingo, Mxico 1969-
LEOBARDO JIMENEZ SANCHEZ, Ph. D. Director, Graduate College, Chapingo, M4xico,
EDWIN J. WELLHAUSEN, Ph. D. Director General CIMMYT 1967-1971.
HALDORE HANSON, Director General, CIMMYT, 1972-1973.
ROBERT D. OSLER, Ph. D. Associate Director, CIMMYT.
JOHN H. LONNQUIST, Ph. D. Head of the Maize Program, CIMMYT, 1967-1969.
ERNEST W. SPRAGUE, Ph. D. Head of the Maize Program, CIMMYT, 1970-1973.
DELBERT T. MYREN, Ph. D. Head of the Communication Department, CIMMYT, 1967-
GREGORIO MARTINEZ VALDES, Ph. D. Head of the Communication Department,
CIMMYT, 1971-1973.
ANTONIO TURRENT FERNANDEZ, Ph. D. Soil Scientist, Maize Program, CIMMYT,
REGGIE J. LAIRD, Ph. D. Soil Scientist, Maize Program, CIMMYT, 1967-1973.

General Coordinator
LEOBARDO JIMENEZ SANCHEZ, Ph. D. Professor of the Graduate College, Chapingo,
Mexico, and member of the CIMMYT staff, 1967-1970.
MAURO GOMEZ AGUILAR, Ing. Agr. 1970-1973.

Technical Staff
ARMANDO PUENTE BERUMEN, Ph. D. In charge of agronomic research, 1967.
J.C. DORANTES DE LA ROSA, Ing. Agr. Associate in agronomic research, 1967.
ANTONIO TURRENT FERNANDEZ, Ph. D. In charge of agronomic research, 1968.
ALVARO RUIZ BARBOSA, Ing. Agr. In charge of agronomic research, 1969-1973.
NESTOR ESTRELLA CHULIN, Ing. Agr. Associate in agronomic research, 1969-1971.
RAMON BARRAZA MADRID, Ing. Agr. Associate in agronomic research, 1971-1973.
JOSE T. MORALES G. Associate in agronomic research, 1973.
MARIO CASTRO GIL, Ph. D. In charge of genetic improvement, 1967.,
HERNAN CORTES MENDOZA, Ing. Agr. Associate in genetic improvement, 1967-1969.

FACUNDO BARRIENTOS P., M.S. In charge of genetic improvement, 1968.
TARCICIO CERVANTES SANTANA, M.S. In charge of genetic improvement, 1969-
FIDEL MARQUEZ SANCHEZ, Ph, D. Geneticist, Graduate College, Chapingo, M4xico,
Special Collaboration in genetic improvement, 1969.
HUGO MEJIA A., Ing. Agr. Associate in genetic improvement, 1969-1971.
APOLINAR AGUILLON GALICIA, Ing. Agr. Associate in genetic improvement, 1971.
In charge of genetic improvement, 1972.
HELIODORO DIAZ CISNEROS, Ing. Agr. In charge of evaluation, 1968-1969.
JUAN MANUEL RANGEL, Ing. Agr. In charge or evaluation 1970.
FRANCISCO GARCIA HERNANDEZ, Lic. In charge of evaluation, 1971-1973.
FELIPE RODRIGUEZ CANO, M.S. Technical assistance to farmers, Zone 4, 1968-1971.
FRANCISCO ESCOBEDO CASTILLO, Ing. Agr. Technical assistance to farmers, Zone 2,
GILDARDO ESPINOZA SANCHEZ, Ing. Agr. Technical assistance to farmers, Zone 1,
J. de JESUS GUERRA MARTINEZ, Ing. Agr. Technical assistance to farmers, Zone 3,
1969; Zone 5, 1970-1971.
ARISTEO ALVAREZ, Ing. Agr. Technical assistance to farmers, Zone 3, 1970.
ROGELIO AGUIRRE RIVERA, Ing. Agr. Technical assistance to farmers, Zone 3,
FELIPE DE LA FUENTE PEREZ, Ing. Agr. Technical assistance to farmers, Zone 1,
ESTEBAN LEAL ZAPATA, Ing. Agr. Technical assistance to farmers, Zone 2, 1971-1972.
EZEQUIEL SOLIS RUIZ, Ing. Agr. Technical assistance to farmers, Zone 5, 1971-1972.
LUCIANO VIDAL GARCIA, Ing. Agr. Associate in evaluation, 1971; Technical assistance
to farmers, Zone 4, 1972.
BENJAMIN BAUTISTA OREA, Ing. Agr. Technical assistance to farmers, Zone 3, 1972.
OZIEL MONTAIEZ GONZALEZ, Ing. Agr. Technical assistance to farmers, Zone 1,
ARTURO CAMARENA MEDRANO, Pas. Ing. Agr. Technical assistance to farmers, Zone
JOSE MARCOS FRANCISCA, Pas. Ing. Agr. Technical assistance to farmers, Zone 2,
MIGUEL SANCHEZ HERNANDEZ, Pas. Ing. Agr. Technical assistance to farmers, Zone
ALFONSO MACIAS LAYLLE, Pas. Ing. Agr. Technical assistance to farmers, Zone 5,
JOSE ZARAGOZA PULIDO, Sociologist, Special collaboration in the organization of
farmers, 1972-1973.
GIL OLMOS BARRERA, M.S., Special Collaboration in Visual Aids, 1968-1973.

Technical Advisors
LEOBARDO JIMENEZ SANCHEZ, Ph. D. Director, Graduate College, Chapingo, Mexico,
REGGIE J. LAIRD, Ph. D. Soil Scientist, Maize Program, CIMMYT, 1967-1973.
DELBERT T. MYREN, Ph. D. Head of the Communications Department, CIMMYT,
ANTONIO TURRENT FERNANDEZ, Ph. D. Soil Scientist, Maize Program, CIMMYT,
GREGORIO MARTINEZ VALDES, Ph. D. Head of the Communications Department,
CIMMYT, 1971-1973.




The physical environment climate, soils. The farming population number of
farmers, size of holdings, type of land tenure, the family and the home, contact
with ideas from outside the community. Local production technology genetic
resources, production practices. Agricultural service agencies.

Evolution of the Project. Project Operations 1967 through 1973.
Coordination. Project Personnel. Consulting Services.

General strategy. Field research in 1967 results: 1967.
Field research in 1968 and 1969 results: 1968 and 1969.
Field research in 1970 results: 1970.
Field research in 1971 results: 1971.
Field research in 1972 results: 1972.
An evaluation of the agronomic research program.

Strategy of genetic improvement. Program and results production of hybrids,
production of varieties, evaluation of materials. Evaluation of the research program.

Project program: 1968 location of the high-yield plots, credit, crop insurance,
planting and care of the high-yield plots, result demonstrations, printed matter and
audio-visual aids. Project program: 1969 organizing groups, radio usage, super-
vision of the high-yield plots, interchanges among farmer groups, results demonstra-
tions. Project program: 1970. Project program: 1971. Project program: 1972.
Project program: 1973 use of agua ammonia, more effective group action.

Organization of the farmers prior to the Puebla Project. Action of the Puebla
Project in the organization of farmers. Benefits received by farmers through organ-
ized action better understanding of the new technology, access of small farmers
to agricultural credit, greater efficiency in obtaining credit, prompt delivery of
fertilizers, efficient and cheaper transport of fertilizers, prompt repayment of loans,
greater efficiency in the repayment of loans, access to information on other activi-
ties, initiation of new production activities, greater effectiveness in solving com-
munity problems, greater interchange of experiences among farmers. Factors favor-
ing group efficiency quality of leadership, legalization of the organization.
Outlook for more advanced forms of organization. Summary.

The Impulsora de Puebla procedure followed in granting credit, changes in sales
of 10-8-4, outlook for greater credit for small farmers from the Impulsora de
Puebla. The Puebla Branch of the National Agricultural Credit Bank operating
procedures of the Agricultural Bank, outlook for greater credit for small farmers
from the Agricultural Bank. The Puebla Branch of the National Ejidal Credit Bank
operating procedures of the Ejidal Bank, outlook for greater credit for ejidatarios
from the Ejidal Bank. Agricultural Bank of the South. The Coxtocan Hacienda.
Direct participation of Guanomex. The National Agricultural Insurance Agency -
operating procedures of ANAGSA, crop insurance related problems that limit
farmer use of official credit. The National Marketing Agency.

Collection of existing information. Personal interview surveys survey: 1967,
survey: 1970. Studies of agricultural institutions and the farm sector. Annual
estimates of maize yields development of an indirect method for estimating
yields, estimation of yields of all farmers in the area, estimation of yields of farmers
on credit lists. Comment on the evaluation program.

Level of adoption of the maize recommendations all farmers in the area, farmers
on credit lists. Changes in average maize yields. Factors influencing the adoption of
the maize recommendations availability of information, adequacy of the new
technology, risk in using the new technology, availability of credit, farmer organ-
izaitons, other factors.

Changes in family income. Changes in employment.
Other changes that influence the general welfare-changes in the consumption of
several foods, improvements in the family home, use of public services, changes in
attitudes of farmers.

Classification of benefits and costs. Estimation of a benefit-cost ratio for seven
years of operation of the Puebla Project direct gross benefits, derived gross
benefits, associated costs, Project costs, adjustment of costs and benefits, benefit-
cost ratios, intangible benefits.

Practical training. Combined theoretical and applied training.

East Antioqufa Project, Colombia. Cajamarca La Libertad Project, Peru. The Basic
grains Program, Honduras. Maize Program, State of Mexico, Mexico. Tlaxcala Proj-
ect, Mexico.

Essential elements of the Puebla approach agronomic research, technical assist-
ance to farmers, evaluations, coordination, technical backing for program staff,
capable, highly-motivated, well-trained staff, incentives, an adequate budget. Pro-
gram strategies influence of the physical environment, influence of infrastructure
development, influence of the political environment, influence of size, diversity and
accessibility of program area, high-yielding crop varieties. The regional agricultural
program a first step in rural development.


117 Some Publications on the Puebla Project


World agriculture faces two problems of great urgency: the threat of an absolute
shortage of food on a global scale, and the fact of continuing low incomes and malnutri-
tion among most of the rural population.
The Puebla Project is an experiment designed to tackle both problems simultaneously
by obtaining a large increase in yield of a basic food crop in this case maize among
small farmers producing at subsistence levels with traditional methods.
Much of the world's food is now produced on small farms, where families produce
mainly for human and animal consumption on the farm and have little or no surplus to
sell. These families have usually been among the last to discard their traditional farming
methods and few of them are quick to reap the benefits of new technology. On a
world-wide basis, however, they represent a vast potential for national development.
This potential is recognized by the governments of most developing countries; but,
because of limited resources and lack of knowledge about how to reach these millions of
smaller farmers, the national programs to increase crop yields have usually been aimed at
a relatively small number of commercial farmers. Yet attention to these families of the
traditional sector is crucial for at least three reasons: (a) their farms represent an
important part 'of the arable land in many countries; thus, yields must be increased to
satisfy total food requirements, (b) in many nations most of the human resources are
employed in traditional agriculture and improved agriculture is a readily available source
of increased capital from within, and (c) traditional farmers make up a large portion of
the population of many countries and continuous improvement in their farming tech-
niques is essential for over-all social development.
These considerations define the need for more efficient means of providing traditional
farmers with better production methods. And it is this need that focused the two initial
objectives of the Puebla Project: (a) to develop, field test, and refine a strategy for
rapidly increasing yields of a basic fbod crop on small land holdings; and (b) to train
technicians from other regions in the elements and successful use of this strategy.


The conceptual framework of the Puebla Project was derived from several interrelated
agricultural and social science disciplines, as well as from the working knowledge and field
experience of the team members, It was conceived as an integrated plan of attack on the
many problems limiting farmer use of adequate production technology. It was assumed
that the following factors of change would need to be available in the Project area: (a)
high-yielding maize varieties, (b) information on efficient production practices, (c) effec-
tive communication of agronomic information to farmers and agricultural leaders, (d)
adequate supplies of agronomic inputs at easily accessible points when they are needed,
(e) crop insurance, (f) favorable relationships between input costs and crop values, (g)
adequate production credit at a reasonable rate of interest, and (h) accessible markets
with a stable price for maize.
When the Project began, several agricultural institutions responsible for providing
inputs, credit, crop insurance and markets for maize producers were already operating in
Puebla. Also, the relationships between the costs of production inputs and the price of
maize were thought to be satisfactory. However, only very limited results were available
from trials of maize varieties and production practices, and only one extension agent was
working the area.

Thus, the action program of the Puebla Project was organized initially to include four
major components: (a) varietal improvement of maize, (b) research to develop efficient
recommendations on maize production practices, (c) assistance to farmers in proper use
of new recommendations, and (d) coordination of the activities of the service agencies,
the Project team, and the farmers. Another component socio-economic evaluation- was
added during the first year.
A key concept within the philosophy of the Project has been that the production and
dissemination of information are parts of a development continuum that should not be
compartmentalized in program operations. Constant interaction among staff members
and feedback of information have been viewed as integral functions of the Project from
planning of research through delivery of findings to farmers and evaluation of results.
Thus, it was planned that the staff consist of a small team of capable, well-trained
scientists with an adequate budget and freedom to operate at any political or technical
level. The team lived and worked in the Project area, cooperating closely in conducting
the field trials, demonstrations, farmer meetings, etc.
The selection and training of team personnel was seen as the crucial element in deter-
mining success. The work of the team was expected to be exceptionally arduous due to
heavy demands by the large numbers of farmers in the Project area. In making decisions,
team members would have to take into account, simultaneously, knowledge and expecta-
tions related to weather, attitudes of farmers, institutional organization, the personal
goals of individuals in key positions, and other factors. Great skill is required in assessing
and giving appropriate weight to these varied and interrelated factors. Thus, strong effort
was given to acquiring the services of well-trained, capable, and innovative young agricul-


Initially, the two conditions considered necessary in selecting the Project area were: an
ecological environment that would permit substantial yield increases, and a political
environment that would be favorable toward Project objectives.
The main requirements of the physical environment were: (a) rainfall and temper-
atures adequate for good-to-high maize yields. The total amount and distribution of
rainfall should be such that maize would suffer severe drought damage in less than 10
percent of the years and moderate damage in no more than 30 percent of the years. There
should be only light frosts, limited to the first quarter of the growing season; and (b)
reasonably deep, permeable soils free from toxic amounts of salts.
The essential aspect of the political environment was that government should strongly
support the Project operations and have the will and the power to modify existing
policies and agencies as necessary. This factor was especially important in respect to
S availability of key inputs, orderly marketing of the grain, and the relationship between
the cost of principal inputs and the price of grain at the farm.
As the Project has evolved, these aspects of the ecological and, to some extent, the
political environments have been recognized, not as prerequisites, but as factors that
influence the strategy to be used in a particular program. The basic approach used in
Puebla should be applicable in most regions of the world, when adequate attention has
been given to the specific environmental, social, and economic conditions in areas where
the approach is to be used.


The area selected for the Project comprises 32 munici-
pios (counties) in the western part of the State of Puebla,
Mexico (Fig. 1.1). In choosing the area, the primary consid-
erations were that it fulfill the ecological and political
prerequisites cited in the Introduction to this report, and
that the farming population should consist mainly of small
land holder.


The Puebla area had two other characteristics that made
it appropriate for the Project: (a) it comprised about
116,800 ha of cultivated land with about 80,000 ha used
for maize production.- it was felt that an area of this size
would be adequate for studying the effectiveness of a new
approach to rapidly increasing maize yields; and (b) pro-
duction practices were traditional, farmer incomes were
low, and most of the harvest was consumed directly on the

Fig. 1.1. The project area covers about 117,000 hectares of crop land in 32 municipios of the State of Puebla. As the Project
evolved, the area was divided into five zones with a technical assistance agent responsible for each.

Communications within the Project area are adequate,
and most villages are connected by a network of all-weather
roads. The Project area is a 2-hour drive from Mexico City
or the National Agricultural Center at Chapingo, and this
ready accessibility allowed consultants at both locations to
maintain close contact with the Project.
The remainder of this chapter describes the physical
environment of the area, the farming population, local
production technology, and agricultural agency services


The Project area occupies much of the valley drained by
the Atoyac River and lies mostly between the rising slopes
of volcanos: Popocat6petl and Ixtaccihuatl to the west, and
La Malinche to the north and east. It is located between
latitudes 18 50' and 190 25' north and between longitudes
970 55' and 98040' west of Greenwich. The lowest part of
the valley lies southeast of the city of Puebla at an elevation
of 2,100 m above sea level. Most of the Project area lies
between 2,150 and 2,700 m above sea level, although maize
is produced on the mountain slopes up to elevations of
2,900 m.


The climate over most of the region is temperate with
mild winters. The warmest part of the year is in May and
early June. Temperatures remain fairly constant during the
last of June, July, and August, and gradually decline during
September and October. Average monthly temperatures
during the maize-growing season vary from 18.60 C in May
to 16.1 C in October.
Frosts occur mainly during the winter months from
October through March, when they cause little or no
damage to annual crops. However, a weather station located
near the center of the Project area reported frost on one or
more days during the month of April in 33 percent of the
years; in May, 17 percent; and in June, 5 percent. Weather
stations at two other locations in the area reported no
frosts during these months. Frosts in May and June can
seriously damage early plantings of maize.
Three of the four weather stations in the Puebla area
reported an average of one hailstorm a month during July
and August, with about half that amount in September.
Severe hailstorms during these months would be expected
to reduce maize yields significantly.
The average rainfall reported by the four weather
stations for the 7-month period from April through
October varied from 777-863 mm. The rainfall during this
period represents approximately 94 percent of the total for
the year.

1. For more information on the climate of the Puebla
valley, see Jauregui, E.O. 1968. Mesoclima de la Regi6n
Puebla-Tlaxcala. Institute de Geografia, Universidad Nacio-
nal Aut6noma de M6xico. M6xico, D.F.

The average rainfall in the Puebla area during the maize-
growing season should be sufficient to satisfy the needs of
the crop. However, drought damage to maize would be
expected when: (a) the total rainfall during the year is
considerably less than the average, or (b) the amounts of
precipitation are well below average during the critical
months of June, July, and August.
In 1967, drought intensities were estimated using
existing information on soil characteristics, evapotranspira-
tion losses, and water needs of maize at critical growth
periods; and the daily rainfall data available at the four
weather stations in the Project area. The drought damage,
estimated for individual years, was classified as: zero or very
slight; moderate; or severe. As an average for the four
stations, it was estimated that there would have been zero
or very slight drought damage in 60 percent of the years,
moderate damage in 30 percent, and severe damage in 10
percent. Maize growing on soils with a high moisture-
supplying capacity would have suffered less from drought
than these percentages indicate, while maize on soils with a
low moisture-supplying capacity would have suffered more.
A moderate effect of drought would be expected to reduce
yields by 30-60 percent; a severe effect by 60 percent or


The soils in the Project area have formed from volcanic
efecta, mainly from the three volcanos: Popocat6petl, Ixtac-
cihuatl, and La Malinche. The parent material ranges in size
from very fine ash to pumice particles several centimeters in
diameter. The coarser materials are found on the upper
slopes of the volcanos and the finer materials near the
center of the valley. The ejecta has probably been water-
reworked over much of the area; some of the ash and
pumice, however, have been deposited directly on the land
surface during eruptions of the volcanos. The parent materi-
als are distinctly layered due to sorting during these
depositional processes.
On the upper slopes of the volcanos the streams are very
deep, and the land surface is being continuously eroded
away. Little of the eroded material, however, reaches the
Atoyac River. Most of the material is deposited as alluvial
fan debris. Alluvial fan building is still occurring in the area
and is especially noticeable along the San Martin Texmelu-
can-Huejotzingo highway, where the stream beds are higher
than the adjacent land surface.
The external drainage system is well-developed on the
upper slopes of the volcanos but is poorly developed
toward the center of the valley where alluvial fans are form-

2. The study of the genesis, morphology, and distribution
of the soils in the Puebla area was carried out during
1968-1970. Dr. B.L. Allen, soil morphologist, Texas Tech
University, Lubbock, Texas, directed and personally con-
ducted much of the field work. He carried out three field
studies, each lasting about a week. Dr. Allen contributed
most of the ideas and information presented in this section.






Fig. 1.2. The distribution of the seven most important kinds of soils in the project area. The regions shown in white are
largely mountainous with little production of maize.

ing. This has resulted in the formation of soils with a high
water table and deficient external drainage on both sides of
the Atoyac River. A drainage system consisting of a net-
work of open ditches was constructed in this region many
years ago and has been effective in maintaining the water
table low enough for the soils to be farmed.

The information accumulated on soils during the early
years of the Project permitted the preparation of a soils
map (Fig. 1.2) showing the approximate boundaries of the
seven most important groups of soils in the Project area.
The total area and cultivated area corresponding to each
soil group were estimated from aerial photographs and are
shown in Table 1.1.

TABLE 1.1. Total area and cultivated area corresponding
to seven groups of soils in the Puebla area.

Total Cultivated
Group area area
ha ha

Deeps soils of Popocatepetl 33,618 26,609
Pumiceous soils of Popocatepetl 26,799 19,794
Soils of La Malinche 34,602 25,298
Sodic-like soils 16,560 13,121
Heavy soils of Zone V 3,151 2,700
Soils with compacted horizon 28,912 22,403
Soils with high water table 7,527 6,871

Total 151,169 116,796

The deep soils of Popocatdpetl cover an area of about
33,618 ha, of which 26,609 are currently under cultivation.
These soils occupy an area extending from the intermediate
slopes of Popocat6petl and Ixtaccihuatl to the soils with a
high water table along the Atoyac River. The predominant
parent material is a light-brown volcanic ash near neutral in
reaction. The texture of the upper 20-40 cm of these soils is
loamy sand or sandy loam. This horizon usually contains
less than 0.5 percent organic matter, has a pH around 6.5, is
high in potassium and calcium, and is fairly high in phos-
phorus. Underlying this horizon is a layer about 2 m deep
with a loam or clay-loam texture that is largely responsible
for the high productivity of these soils. This lower horizon
has a pH of about 7.0, a cation exchange capacity of
around 15 milliequivalents/100 g of soil, and 8 percent of
available moisture when wet to the field capacity. These
soils, when properly managed, can be planted in April and
early May with residual moisture from the previous year.
The pumiceous soils of Popocatdpetl cover an area of
some 26,799 ha, of which 19,794 are currently under
cultivation. The parent material of these soils is largely
pumice with some volcanic ash. According to Aeppli and
Schoenhals3, these materials were deposited at the time of
the most recent eruptions of Popocat6petl. The stratifica-
tion observed in most of these soils probably resulted from
differences in the materials deposited at different times, not
from pedogenic processes.

3. Aeppli, Hans and Schoenhals, Ernst, 1973. Los suelos
de la Cuenca Puebla-Tlaxcala. Comunicaciones 7/1973
Proyecto Puebla-Tlaxcala. Eds. Wilhelm Lauer and Erd-
mann Gormsen. Fundaci6n Alemana para la Investigaci6n
Cientifica, M6xico. pp. 15-18.

The texture of the upper horizon of the pumiceous soils,
which may extend to a depth of 80 cm, is gravelly sand.
Despite its coarse texture, this horizon contains about 6
percent available moisture when wet to the field capacity.
It usually contains less than 0.5 percent organic matter, has
a pH of about 6.5, and a cation exchange capacity of
around 6 meq/100 g. The lower horizon may be similar to
the corresponding horizon in the deep soils of Popocatdpetl
or it may be a loose, pumiceous gravel. This latter material
has a pH near 7.0, a cation exchange capacity of around 7
meq/100 g, and contains about 6 percent available moisture
when wet to the field capacity. When properly managed,
the pumiceous soils can be planted with residual moisture.
The soils of La Malinche cover some 34,602 ha with
about 25,298 ha under cultivation. These soils have formed
from volcanic ash that has been redeposited by water in
alluvial fans. The upper layer of these soils, about 30 cm in
depth, is usually a sand. This horizon contains 0.5-1.0 per-
cent organic matter, has a pH of around 6.5, is high in
phosphorus and potassium, and contains about 7 percent
available moisture when wet to the field capacity. A gravel-
ly sand with a depth of a meter or more is usually found
underlying this horizon.
Soils with an incipient B horizon are found on the inter-
mediate slopes of La Malinche. This B horizon has a loamy
sand to sandy loam texture and a cation exchange capacity
of about 15 meq/100 g, with a low base saturation percent-
age (around 35). On the lower slopes of La Malinche, the
lower soil horizon consists of sediments a meter or more in
depth, with a sandy loam to silty clay loam texture. These
soils are potentially very productive. When properly man-
aged, the soils of La Malinche also can be planted with
residual moisture.
The sodic-like soils occupy an area of about 16,560 ha,
with 13,121 ha currently under cultivation. The parent
material is a light-grey volcanic ash, alkaline in reaction,
which is found essentially unaltered at a depth of 60-160
cm, depending upon the degree of weathering of the
profile. These soils are similar morphologically to Solonetz
soils. The surface or A horizon is a loamy sand about 20 cm
deep. This overlies a Bt horizon some 60-100 cm in depth,
which is black, contains around 60 percent clay, has a
columnar structure, and has a very firm consistency. This
horizon is very impermeable and greatly restricts the pas-
sage of water and the penetration of maize roots. A thin,
greyislf-colored A2 horizon showing the effects of reduc-
tion processes, is usually distinguishable between the A and
Bt horizons.
The Bt horizon has a cation exchange capacity of about
35 meq/100 g, a base saturation percentage of 80, and is
slightly alkaline. The content of exchangeable sodium varies
from 4-14 percent of the cation exchange capacity; thus,
the soils are designated sodic-like rather than sodic. Suf-
ficient moisture to permit early plantings usually cannot be
conserved through the winter months. The production
potential of these soils is very low under rainfed conditions,
but relatively high when irrigation water is available.

The heavy soils of Zone V cover an area of some 3,151
ha, of which 2,700 ha are under cultivation. The morphol-
ogy of these soils is similar to that of the sodic-like soils,
except that the Bt horizon is less developed and the soils
thus have better internal drainage. The heavy soils of Zone
V have a limited capacity to conserve moisture over the
winter months; thus, maize plantings usually can not be
made until the rains begin.
The soils with a compacted horizon occupy an area of
about 28,912 ha, with 22,403 ha presently under cultiva-
tion. The compacted layer that limits both root penetration
and water percolation lies at a depth of 20-60 cm from the
surface. This layer can be: (a) fragipan, (b) claypan, or (c)
tepetate. The first two layers are genetic horizons; tepetate
is a partially consolidated volcanic ash. The surface horizon
contains 0.5-1.0 percent organic matter, has a pH of around
6.5, is high in potassium, and contains moderate amounts
of phosphorus. Sufficient moisture to permit early plan-
tings of maize cannot be conserved through the winter in
these soils.
The soils with a high water table cover an area of some
7,527 ha, of which 6,871 ha are under cultivation. The
depth of the water table in these soils varies with their
proximity to the Atoyac River, with the seasons, and with
the upkeep of the drainage canals. High yields of maize and
alfalfa are obtained in soils with the water table more than
50 cm below the surface. These soils are dark-colored
loams, high in organic matter, with little profile develop-
ment. They have pH values around 7.5, cation exchange
capacities of about 30 meq/100 g, and contain less than 5
percent exchangeable sodium.


Most of the farmers in the region are descendents of the
Indian populations present in the area at the time of the
Spanish conquest. In certain villages, Nahuatl, or "Mex-
ican" as it is known in the area, is still spoken, although
everyone understands Spanish. The information presented
in this section comes mainly from personal interview
surveys for 1967 and 1970 (refer to Chapter 8).

Number of Farmers

The number of farm operators in the Project area was
estimated to be 43,300. This value was calculated by
dividing the total cultivated area, 116,800, by 2.7, the aver-
age number of cultivated hectares per farm operator. This
latter value is the average of the estimates of the number of
cultivated hectares per operator from the 1967 and 1970

According to the surveys, the average family consisted of
5.54 members in 1967 and 6.17 members in 1971. As-
suming there was no change during the 4-year period in the
number of farms in the area, this means the total popula-
tion included in the families of farm operators was approx-
imately 240,000 in 1967 and 267,000 in 1971.

In addition, there are many families living in the villages
who do not operate farm land but depend heavily on
agriculture for employment and sustenance. Using 1960
Census data for the 32 municipios that comprise the Project
area, and the above estimate of the total population of the
families of farm operators, there were estimated to be
13,300 landless rural families in the area in 1967.

Size of Holdings

The average farm size, as mentioned earlier, was esti-
mated to be 2.7 ha. The amount of land per farm operator
varied considerably as shown below.
Amount of Land Operated Percentage of Farmers

.50 ha or less 9.2
.51 1.00 17.5
1.01 1.50 18.7
1.51 -2.00 10.7
2.01 2.50 12.3
2.51 3.00 8.4
3.01 3.50 4.4
3.51 4.00 6.4
4.01 5.00 4.4
5.01 7.50 4.0
7.51 10.00 2.8
10.00 or more 1.2

Type of Land Tenure

The prevalence of different land-holding systems in the
area is indicated by the following data from the 1967

F -%f i 'I A -I )m6NsaHHllrllwllllll

Percent of Hectares
Land Farm Operated Percent of
Holders Operators Per Farmer Total Area

Ejidatarios 38.2 2.05 31.9
Private holders 27.5 2.74 30.3
Ejido-private 33.5 2.76 37.2
Rented 0.4 3.00 0.5
On shares 0.4 0.20 0.1
100.0 2.47 100.0

Of special interest is the frequency of combined private
and ejidal holdings. Ejidatarios are farmers who received
land from the government as a result of the agrarian reform.
They have the use of the land while they live, but they
cannot sell it. Also, private holders value their land very
highly and sell only under exceptional circumstances.
Consequently, the land market in the area is essentially

Farmers commonly have several parcels at various loca-
tions on different kinds of soils and at varying distances
from the farmstead, as shown below.

No. of Parcels Percentage of Farmers
per farmer

1 16.7
2 or 3 52.5
4 or 5 21.2
6 or 7 5.6
8 or 9 4.0
10 or more 0.0
HK *I..UL;^rf ^t~t S5SrE~r .y-WKKlSCWS~s ^.t

The average farm size in the project area is about 2.7 ha. Approximately 90% of the farms have 5 ha. or less.

At certain peak labor periods,
such as planting and harvest,
the whole family helps in the
field and additional workers
may be hired. As the fields
are often far from the village,
the women usually bring the
noon day meal.

This phenomenon is explained in part by the farmers'
awareness of land quality. To be fair to all when the ejidos
were organized, the farmers frequently decided that each
ejidatario should have a piece of each of the two or three
qualities of land, instead of a single block of land. A similar
procedure is often followed by private owners in leaving
land to their children.

The Family and the Home

The family provides both the management of the re-
sources used in agricultural production and most of the
labor used on the farm. Where labor is hired, it is usually
for short periods of peak activity, such as at harvest time
when there may, in fact, be a shortage of labor in the
region. Much of this hired labor is offset by members of the
family working off the farm at other periods when labor
needs on the farm are low.
About 77 percent of the farm operators have some
ability to read and write, as shown in the following data
from the 1967 survey.

Schooling Percent of Farm Operators
Never attended school:
Illiterate 22.7
Self-taught, literate 4.3
Attended school for:
I year 10.0
2 years 17.1
3 years 22.3
4 years 9.6
5 years 4.4
6 years 8.0
More than 6 years 1.6
Average years of schooling:
All farmers 2.4
Farmers who attended school 3.2

The impression is that at least half of those with some
degree of literacy read and write with considerably dif-
ficulty. The percentage of farmers with 3 years or more of
schooling, 45.9, is perhaps a reasonable estimate of those
that are functionally literate.
It is customary in the area for farmers to live together in
villages. Homes are usually (76 percent) made of sun-baked
adobe bricks. The floors are frequently of brick, cement, or
tile, but 36 percent are of dirt. The dwellings are small as
indicated in the next table (1967 survey).

Rooms (other than Kitchen)
in the Farm Home Percent

One which is also the kitchen 1.2
One room 43.8
Two rooms 32.3
Three rooms 14.7
Four rooms 5.2
Five or more rooms 2.8

Though most of these families live humbly, many have
some of the minimum comforts associated with modern
living as can be seen in the following table (1967 survey).

Living Comforts Percent
Have electric lights 62.9
Have radio 59.8
Have a sewing machine 45.0
Cook with gas, electricity, or fuel oil 28.7
Have piped water in home or street 13.1
Have television 8.0
Have drainage 6.0
Have refrigerator 1.6


In nearly every case, the family diet depends heavily on
home produced food. The poorest families eat practically
nothing besides maize and beans with small quantities of
chiles, onions, and tomatoes for seasoning. Those with
more resources occasionally consume wheat bread, eggs,
and meat, and their children drink milk.
The average total family income, estimated from the
information provided by the 251 farmers in the 1967
survey, was $666.80. As is shown below, the income came
from four main sources:

Sources of Family Income Percent

Net income from crops 30.4
Net income from animals 28.4
Off-farm wage income 23.7
Other non-farm income 17.0

Although most of the family income was derived from
farming activities, there was some non-farm income, mainly
from domestic, commercial, and industrial employment in
nearby cities.

Contact with Ideas from Outside the Community

There exists an excellent opportunity for contact with
the large urban society outside the villages. Local roads are
rough and eroded, but in most cases they are passable
during the entire year. The local buses are battered with
years of wear, but provide a regular and inexpensive means
of transportation for both people and produce. Traveling
outside the village, however, is not undertaken casually.
According to the 1967 survey, only 24 percent of the farm-
ers leave the village at least once a week. Another 14
percent leave every 2 weeks or every month, 43 percent
rarely leave the village, and the remaining 19 percent
reported that they never leave the village.
In spite of limited physical mobility, there is contact
with ideas from outside the villages, principally through
radio, as suggested by the following data from the 1967

Percent of farmers
Have a radio 59.8
Listen to it daily 50.2
Listen to a farm program 21.9
Have a television set 7.9
See TV at home or elsewhere
at least once per week 12.4
Read farm magazines regularly 1.6
Read newspapers regularly 7.9


The percentages of the cultivated land used for different
crops were estimated from survey data for 1967 and 1970:


Percent of
Cultivated Area

Maize 69.4
Beans 15.9
Alfalfa 5.3
Vegetable crops 3.0
Fruit trees 0.9
Others 5.5

Most of the maize, beans, and fruit trees are produced
under rainfed conditions. Alfalfa and vegetable crops
usually receive supplementary irrigation.
The important cropping systems under rainfed condi-
tions in the Puebla area are: (a) maize alone, (b) the
maize-pole bean (Phaseolus vulgaris) association, (c) bush
beans (Phaseolus rulgaris) alone, (d) maize interplanted in
orchards, and (e) scarlet runner beans (Phaseolus coc-
cineus), known locally as avocote. Pumpkin (Cucurbita
spp.) is interplanted in many of the maize fields with a
population density of 300-1,000 plants/ha.
Maize planted alone is the most important crop in the
Project area. The maize-bean association is the second most
important cropping system in Zones I, II, and IV (see Fig.
1.2). Bush beans is the second most important crop in Zone
III. Bush beans and avocote follow maize in importance in
Zone V, yet occupy a small fraction of the total area. Bush
beans is the third most important crop in Zones I, II, and
IV. Maize interplanted in orchards is most common in Zone

^ 3

The staple food of the rural families in the Puebla area, as
in the rest of Mexico, is maize. It is eaten mainly in the
form of a thin bread or tortilla. The average annual con-
sumption of maize per person in the Puebla area is about
250 kilos.

The local production technology for these crops is the
product of centuries of interaction among the farmers, their
environment, and external influences. There is solid evi-
dence that primitive wild maize was domesticated as long
as 7,000 years ago in the highland region of which the
Project area is a part. When the Spanish conquerors arrived
at Cholula (near the center of the Project area) they found
the inhabitants cultivating maize. The historian Bernal Diaz
del Castillo reports that the farmers in the valley of Mexico
at the time of the conquest were using human excrement
and fish bones to fertilize maize. It is probable that knowl-
edge of these practices extended to the nearby Puebla

.Genetic Resources

Great phenotypical diversity is evident in the local
varieties of maize, bush beans, pole beans, and pumpkin.
Most of the local maize varieties belong to the Chalquefio
race. For early plantings in March and April, farmers use
late-maturing varieties that flower in 100-120 days and
require about 180 days to reach physiological maturity.
These early plantings usually experience some moisture
stress during the first 2 or 3 months, but have adequate
moisture during the rest of the growing season. The late
varieties are generally high-yielding; yields of 10 ton/ha of
grain have been reported in field trials.

Farmers use early-maturing varieties with a biological
cycle of about 130 days for late plantings in June. These
varieties flower in 75-90 days after planting. Typical
environmental conditions during the early growth stages of
late plantings are cool temperatures, low light intensities,
and abundant moisture. The yielding potential of early
varieties is only about half that of late varieties.

In addition to this relationship between earliness and
yielding potential, the length of the growing cycle of maize
varieties tends to be correlated with the height of the
plants, shape, texture, and color of the grain, and (prob-
ably) tolerance to early drought. Late-maturing varieties are
about 3 meters tall and have grain that is usually dented,
hard, and light-colored. Early-maturing varieties are about 2
meters tall and produce large kernels that are usually floury
and dark-colored (red or blue).

Maize varieties with an intermediate growing season also
are available in the area for May plantings. Most native
varieties are susceptible to lodging when produced under
favorable growing conditions.

Data collected in the 1967 survey indicated that 54.6
percent of the farmers knew about hybrid varieties of
maize. About 15 percent of the farmers had planted hybrid
maize on at least one occasion, but only 0.8 percent of
them planted a hybrid in 1967. Of the farmers who knew
of hybrid maize but had never planted it, 64.2 percent gave
as their reason that hybrids did not outyield their local
varieties, or did so only under irrigation.

The length of the growing season of beans tends to be
correlated with the growth habit and the size, form, color,
and flavor of the grain. Beans with a long growing season
have an indeterminate growth habit (pole beans), large
grains, light colors, and a flavor preferred by most con-
sumers. Beans with a shorter growing cycle have a deter-
minate growth habit (bush beans), smaller grain, and a less
popular flavor. (The yielding potential of the maize-pole
bean association is suggested by the results from a later
field trial in which the association receiving both chemical
fertilizers and chicken manure produced 4.5 tons of maize
and 3.0 tons of beans per hectare).
Little is known at present about local production tech-
nology for ayocotc and pumpkin. The experimental study
of the maize-pumpkin association was not undertaken until

Production Practices

Most farmers manage their soils so that moisture present
in the profile at harvest time is conserved through the
winter months; thus maize can be planted in March and
April, some 2 months before the rainy season begins. Farm-
ers cut and shock their maize as soon as it reaches phys-
iological maturity; they then plow immediately and smooth
the surface with a wooden plank. This operation is usually
done in October and repeated in February or March. Maize
planted in these soils with residual moisture usually suffers
from drought before the summer rains begin. This moisture
stress slows down or stops vegetative growth, but other
physiological processes continue. As soon as the rains begin,
the maize continues its vegetative development. If drought
is not too prolonged, the maize has sufficient time to
produce large plants and a good yield. By preparing their
land in this manner, farmers (under rainfed conditions) are
able to use late varieties that require 180 days to reach
maturity in an area where the period with rainfall and
without critical frosts lasts only 140 days.
Farmers who plant with residual moisture understand
that agronomic risk in their plantings is due primarily to
drought during the period between planting and the begin-
ning of the rainy season and to the midsummer or intra-
estival drought (usually between July 15 and August 15).
Those who plant early are betting that drought during the
period before the rains begin will not be severe; they stand
to gain a high yield if early drought is slight. Farmers who
plant late are betting that early drought will be severe; they
stand to gain yields less than those produced by early
plantings if early drought is slight-but will produce com-
paratively higher yields if early drought is severe. Thus,
planting date is a variable that can be manipulated. The
usual practice is for farmers to use a mixed strategy in
choosing the date of planting; that is, they distribute early
plantings over a period of a month or so.
Present technology does not provide for the conserva-
tion of sufficient moisture during the winter months to
permit early plantings of maize in sodic-like soils, heavy

soils of Zone V, and soils with a compacted horizon. These
three groups of soils occupy 33 percent of the cultivated
land in the Project area (Table 1.1.. In addition, some of
the farmers with soils capable of retaining sufficient mois-
ture for early plantings are unable to plant early because
they do not employ moisture conservation practices at the
proper time. Furthermore, it is usually not possible to
conserve sufficient moisture to permit early maize plantings
in soils where maize is interplanted in orchards. These three
categories account for a sizeable area that is not planted
until the rains begin. Farmers use short-season varieties for
these late plantings. In the rare years when rains do not
begin until early July, farmers prefer not to plant maize
because of the frost hazard. According to 1967 survey data,
maize plantings for that year were spread over a three
month period, as shown below.

Date of Planting Percentage of Land

Before March 1 3.8
March 1-31 34.6
April 1-30 37.4
May 1-31 17.8
After May 31 6.4

In their plantings of maize alone or in association, farm-
ers use between 15,000 and 35,000 plants/ha, depending
upon the fetility level of the soil and the amount of ferti-
lizer to be used. If pole beans are grown with maize, the
farmers sow sufficient seed on the same date to give a bean
density of 5,000-20,000 plants/ha. A common rule-of-
thumb is one bean plant for every two plants of maize.
Maize plants that have no adjoining bean plant help support
the weight of the beans and prevent lodging. The distance
between rows of maize is about 90 cm.
Farmers use two methods for planting and cultivating
bush beans: (a) planted in rows 70 cm apart and cultivated
with animal-drawn implements, and (b) planted in rows 50
cm apart, using the method called a barbecho. In this

method, the soil is turned with a single-moldboard plow,
the seed is deposited at the bottom of the furrow, and then
it is covered by the following passage of the plow. Cultiva-
tions are made with a hoe. Population densities are about
60,000/ha for the first method; 90,000/ha for the a bar-
becho method. The a barbecho method is commonly used
in soils with a high content of pumice in the plow layer,
possibly because of the ease of hand weeding.
According to the 1967 survey data, 95.2 percent of the
farmers in the Project area knew of chemical fertilizer; 80.1
percent had used it on at least one occasion; and 69.3 per-
cent had used it in 1967. These farmers in 1967 used an
average of 49.3 kg/ha N and 20.3 kg/ha P205. For the
entire cultivated area in 1967, the average amounts were 34
kg/ha N and 14 kg/ha P205. Of the farmers applying ferti
lizers in 1967, 64 percent used the formula 10-8-4, 18 per-
cent used ammonium nitrate or ammonium sulfate, and 15
percent used other formulas.
Chemical fertilizers have been used for several years in
the Puebla area as indicated in the survey data shown next.

In What Year Did You
First Use Chemical Fertilizers? Percentage of Farmers

Have never used it 19.9
1967 7.5
1966 7.5
1965 10.4
1964 6.4
1963 6.4
1962 6.0
1961 2.0
1960 6.8
1959 or before 27.1

When the Project began in
1967, 95%of the farmers
knew of chemical fertilizer
and 80 % of them had used it
at least on one occasion. Most
felt that some fertilizer was
needed, but did not know
which elements or how much
to apply.

Of the 251 farmers interviewed in 1967, 191 provided
information on time of applying fertilizers to maize as
shown next. Most farmers applied fertilizer only once, at
the time of the first cultivation.

Time of Fertilizer Application Percentage of Farmers

At planting 2.6
At first cultivation 63.4
At second cultivation 24.1
At both 1st. and 2nd
cultivations 9.9

Maize planted alone or m association is usually cultiva-
ted twice: the first cultivation 30-40 days after planting,
and the second cultivation 60-70 days after planting. Farm-
ers take into account the amount of soil moisture, prob-
ability of frost, and size of the weeds, in deciding when to
make the first cultivation in early plantings of maize using
residual moisture. Cultivating is postponed if a frost is
thought likely. The second cultivation is made after the
maize is 60 days old and when the soil is moist to a depth
of at least 30 cm. Some weeds are allowed to grow in the
maize and are harvested for forage as needed.
Farmers in Zone III practice a 2-year rotation of maize
and bush beans. Chicken manure (5-10 tons/ha) is applied
to the maize immediately before the first cultivation. Bush
beans are then grown the following year without fertiliza-
tion. Yields of both maize and beans are relatively high
using this rotation.
It is a common practice in Zones I and V (and to a lesser
extent in the rest of the area) for farmers to top the maize
plants when the grain is still in the milk stage and to cure
the tops in the field and preserve them as high-quality
forage. The rest of the plant is cut and shocked as soon as
the grain reaches physiological maturity. Later, the ear is
harvested and the rest of the plant is preserved as forage of
a poorer quality than that of the tops.
Farmers use few insecticides and herbicides. Forty-one
percent of the farmers in 1967 knew of chemical insecti-
cides and 22 percent had used them on some occasion,
principally to control insects on beans.


Two official credit banks, the National Agricultural
Credit Bank and the National Ejidal Credit Bank, were
operating in Puebla in 1967. According to survey data, 6.4
percent of the farmers received credit from the Ejidal Bank
that year and 0.4 percent from the Agricultural Bank.
Another 5.2 percent of the farmers obtained credit from
other sources, mainly private lenders. A third official bank,
the National Crop and Animal Production Bank, established
a branch in Puebla in 1967.

Only 39% of the farmers sold maize in 1967. Most of this
maize was sold to local buyers who picked it up at the farm
home, or it was retailed in the village on market days. The
maize was marketed throughout the year, to cover various
costs, especially medical care.

Crop production inputs-fertilizers, improved seeds,
insecticides, herbicides, etc.-were available to farmers in
1967 through merchants in the principal cities of Puebla,
San Martin Texmelucan, Huejotzingo, and Cholula. In
addition, there were 42 villages in which a total of 80 store
keepers bought fertilizer and resold it at the local level.
According to survey data, only 38.8 percent of the farm-
ers sold maize in 1967. This maize was marketed through-
out the year, mainly to cover the costs of medical care.
About 30 percent of the maize was purchased by the
National Marketing Agency, CONASUPO, at the support
price of $75.20/ton. The remainder of the maize was sold
to local buyers at the official price or slightly less.
A branch of the National Agricultural Insurance Agency
operated in Puebla in 1967 and provided crop insurance to
farmers receiving credit from the official banks. An agency
of the National Agricultural Extension Service was located
in the city of Puebla, with one extension agent in San
Martin Texmelucan. There was no agricultural experiment
station, but investigators of the National Agricultural
Research Institute occasionally conducted field trials in the
Project area.
The characteristics of the service agencies, their
activities during 1967-1973, and factors limiting their
effectiveness are discussed in Chapter 7.




The Project area was selected in early 1967 after evalu-
ating information collected in visits to the area just prior to
the maize harvest in 1966, reviewing the results from earlier
experiments, and studying weather data from several loca-
tions in the region. Members of the CIMMYT staff played a
key role in the initial planning and setting up of operations.
Financial support for the Project was approved in March,
1967, and a research agronomist and a maize breeder were
employed soon afterwards. The first field experiment was
installed on April 18, 1967.
This chapter is designed to highlight program operations
as they evolved from 1967 to 1973, and to briefly describe
the coordinating and staffing functions.


Several agricultural service agencies were operating in the
Puebla area when the Project was organized, as indicated in
Chapter 1. The role of the Puebla Project, as conceptualized
by its planners, was to complement the activities of these
existing agencies by (a) providing those services that were
lacking, and (b) coordinating the total effort to assure
adequate and accessible agricultural services for the small

Project Operations: 1967

The information available in early 1967 indicated that
maize yields would be greatly increased by applying
adequate quantities of nitrogen and phosphorus, using
higher plant densities, and controlling weeds properly.
Thus, the emphasis in the first year was on agronomic
research to identify outstanding maize varieties and to
determine optimal packages of production practices. The
program to provide technical assistance to farmers was
postponed until reliable recommendations on maize pro-
duction practices were available.
Although the search for a Project coordinator was begun
in March 1967, four months elapsed before a qualified person
was found for the position (detailed discussion of the
coordinator's role is reserved for following section). On
joining the Project in August 1967, the coordinator carried
out a general reconnaissance of the area, traveling over most
of the all-weather roads, and observing the technology used
by the farmers in maize production. Farmers were ques-
tioned about their production practices, average yields,
relationships with agricultural institutions, and possible
interest in participating in the Project.

This exploration led to a better understanding of the
nature of the problem of increasing maize production in the
Project area. It brought clearly into focus the need for a
formal survey to gather more detailed information on the
characteristics of the farmer and his family, the production
technology in use at the time, farmers' attitudes toward
change, etc. As a result, the decision was made to add
socioeconomic evaluation as another component of the
Project. An evaluation specialist was selected in late 1967,
and the first personal interview survey of the farmers was
made in January and February of 1968.
Concurrent with the study of the Project area and the
farming population, the coordinator proceeded to con-
solidate relationships with the agricultural institutions in
Puebla. Interviews were held with the leaders of the dif-
ferent institutions, and they were informed about the
philosophy, objectives, organization, and operation of the
Project. Discussion in these interviews emphasized the
importance of the role of each institution in achieving the
goals of the Project.
The interviews also allowed the coordinator to become
familiar with all national, state, and local institutions, as
well as private organizations, involved in agricultural devel-
opment in the area. A study was made of the objectives,
organization, and operating procedures of each institution,
and an understanding was sought of the decision-making
process and the responsibilities of key individuals of the
different organizations. This knowledge was helpful in
deciding how to proceed in seeking a solution to particular
The fertilizer and maize breeding experiments were
harvested as soon as the maize reached maturity. Then the
results were analyzed and a general recommendation was
formulated for producing maize in the area. This recom-
mendation called for a fertilizer treatment of 130-40-0, a
plant density of 50,000/ha, early control of weeds,
chemical control of high infestations of rose chafer at
flowering, and the use of native maize varieties.
The first Annual Meeting of the Puebla Project was held
at Puebla in December 1967. Representatives of all agricul-
tural institutions were invited, and the Project staff ex-
plained the experimental findings and the maize recom-
mendation for 1968.

Project Operations: 1968
The package of recommended practices for 1968 implied
three principal changes for the agricultural institutions: (a)
an increase in the amount of credit per hectare needed to
purchase fertilizers, (b) substitution of ammonium sulphate

and ordinary superphosphate for the formula, 10-8-4, and
(c) availability of the credit and fertilizer materials at the
local level in March rather than in May.
The changes implied by the new recommendation were
discussed individually with representatives of the different
institutions. In general, it was found that the institutions
accepted the findings of the Project, but were uncertain as
to their participation. The three official banks were reluc-
tant to introduce changes of the suggested magnitude until
their value had been demonstrated in a network of com-
mercial plantings. The banks felt their credit programs to
farmers were satisfactory. In their view, the changes recom-
mended by the Project would present additional risks
because of the larger amounts of credit required, as well as
a possible loss of prestige should the new technology not
provide good results.
The crop insurance agency maintained that, after the
new recommendation had been accepted by the credit
banks and their clients, the farms using the recommenda-
tions could qualify for insurance. This agency, however,
had rigid operating procedures that did not permit coverage
for individual small farmers. Again, change was necessary if
the crop insurance agency were to participate; these
changes required the presentation of proposals to higher
authorities and favorable action at that level.
The Project plans drafted in early 1968 called for a
continuation of research on maize improvement and agron-
omic practices, and the initiation of the program of tech-
nical assistance to farmers. The assistance program was to
consist of approximately 100 "high-yield" plots of 0.25-1.0
ha. The farmers owning these plots would use Project
recommendations, and they would be supervised by a tech-
nical assistance agent who had been added to the Project
staff in early 1968. The experiments and high-yield plots
were limited to the western two-thirds of the Project area
(known later as Zones I through IV) in order to concentrate
the efforts of the available staff.

At the end of each season,
after results had been evalu-
ated, meetings were held at
which the members of the
Project team presented their
findings and recommenda-
tions to representatives of the
agricultural service institu-

The cost of the fertilizer for the high-yield plots could
have been considered a demonstration cost and paid by the
Project. However, because the plots of 0.25-1.0 ha rep-
resented a large portion of the total area in maize of many
farmers, it was decided that free provision of fertilizer
would establish a difficult precedent. Also, procedures for
obtaining credit required immediate testing, to facilitate
channeling of credit and fertilizers to more farmers in the
Two of the official credit banks decided not to provide
credit for the high-yield plots in 1968. The third official
bank agreed to finance about 20 percent of the high-yield
plots. Another 20 percent of the plantings were financed by
the farmers. The remaining 60 percent were financed by a
private fertilizer distributor, Agr6nomos Unidos, which
made the fertilizer available on credit at an interest rate of
1'/2percent/month(these loans were guaranteed by CIM-
MYT). A total of 141 plots were provided for, far in excess
of the 100 plots originally planned.
Field personnel of the credit banks provided suggestions
for locating farmers to cooperate in the Project in 1967 and
early 1968. In conducting the socioecor.omic survey in
early 1968, however, the evaluation team encountered
negative reaction, and sometimes open hostility, from the
farmers in several villages. This experience, plus other obser-
vations in the area, strongly indicated that contact should
be made directly with the local authorities of each com-
munity. The coordinator proceeded to establish contact
with each village in the area and to hold meetings to explain
the objectives and operation of the Project and assess the
interest of the farmers.
The experimental plantings and the high-yield plots were
used as demonstrations of the importance of the improved
production practices from the time the maize began to
flower until harvest. Throughout this period, field days were
held for representatives of the agricultural institutions and
for groups of farmers. The field days for institutions had



two principal objectives: (a) to convince the leaders of the
institutions that recommendations, based on the field
experiments, represented the most reliable information
available for increasing yields, and (b) to acquaint these
leaders with the capability of the Project staff. The field
days for farmers sought to demonstrate the results they
could expect through use of the recommended practices.
During 1968, audio-visual materials were prepared, using
the results obtained in 1967 and the experimental plantings
and high-yield plots. These were prepared specifically for
use in reaching large numbers of farmers in subsequent
Just prior to harvest in 1968, the evaluation specialist
determined grain yields of the high-yield plots and of a
sample of farmers' plantings. Experimental plantings were
harvested in October and November, the data were anal-
yzed, and new maize recommendations were proposed. The
Second Annual Meeting was held in December 1968, to
inform representatives of the agricultural institutions of the
accomplishments during the year.
By the end of 1968, several findings were clear: (a)
large increases in maize yields could be obtained through-
out the Project area; (b) after seeing the results of agronomic
research, the technical assistance to farmers, and the evalua-
tions, representatives of the agricultural institutions were
convinced of the value of Project recommendations; (c) the
farmers who had cooperated with a part of their land in
high-yield plots were prepared to assist other farmers in
using the new technology; and (d) the Project staff, farm-
ers, and agricultural agencies could be effectively coordi-
nated in working to achieve the goals of the Project.

Project Operations: 1969

Plans for 1969 called for three major modifications in
the operation of the Project: (a) research activities would
be extended to cover the entire area; (b) the western three-

-A '
"L'a-"' *^{.'^..ii~t JtI

fourths of the area would be divided into four zones (see
Fig. 1.2), and a technical assistance agent would be assigned
to each zone; and (c) a coordinated effort would be made
to enlist 5,000 farmers to use Project recommendations on
10,000 ha of maize.
The decision to promote the use of Project recommenda-
tions so extensively in 1969 was made after careful study
and discussion by the Project staff and state representatives
of national agricultural institutions. Expansion of the pro-
gram to reach 10,000 ha of maize would require credit
needed for fertilizer alone amounting to about $560,000.
Discussions with the different institutions led to the pro-
posal that the National Ejidal Credit Bank would finance
2,000 ha, the National Agricultural Credit Bank 1,500 ha,
the Agricultural Bank of the South 1,500 ha, and that the
Impulsora de Puebla, through its subdistributor, Agr6no-
mos Unidos, would finance 5,000 ha. This proposal was
approved at the national level.
The plan to reach 5,000 farmers in 1969 implied a
drastic change in operating procedures of the technical
assistance agents. In 1968, one technical assistance agent,
assisted by the evaluation specialist, had given individual
attention to 103 farmers. In 1969, four agents would have
to assist up to 5,000 farmers. Such assistance seemed pos-
sible only if the farmers organized into groups.
The four technical assistance agents launched a program
to inform the farmers in the area about the maize recom-
mendations, and to assist them in organizing into groups
and arranging for credit and fertilizers. The field personnel
of the credit banks, the National Marketing Agency, and
the crop insurance agency, assisted in this work. In their
relationships with the farmers, the technical assistance
agents made sure that arrangements for credit and fertilizers
were always made between the credit institutions and the
leaders of the farmer groups. Thus, the responsibility for
procuring credit and fertilizers remained with the individual
farmer or the group leader.

During the cropping season,
field days were held for farm-
ers and representatives of the
agricultural institutions. Here
the Minister of Agriculture,
the Governor of Puebla and
other dignitaries listen to an
explanation of the agronomic
research program.

Perhaps the most crucial period in the development of
the Project was in early 1969. Although the specific func-
tions of the Project staff and participating institutions were
defined well in advance of planting, the task remained of
coordinating all activities so that the farmers would have
fertilizers when needed. The correct fertilizer materials had
to be ordered early, freight cars had to be available to trans-
port the material to Puebla, the shipments had to be
received by the distributors and dispatched to subdistri-
butors, and the farmers had to arrange for credit so they
could pick up their fertilizers when needed. Problems
developed at all points in this chain of events, and continu-
ous contact by the coordinator and a clear understanding of
the operation of each institution provided the means to
reduce delays to a minimum and to avoid the loss of
prospective cooperators.
A total of 2,561 farmers, organized in 128 groups, were
assisted in using Project recommendations on 5,838 ha.
Although the total fell short of the stated goal for 1969,
some 5.4 percent of the farmers in the area did begin using
the new technology that year.
During the fall of 1969, regional demonstrations were
held at six locations in the area. The average maize yields of
organized farmers and all farmers in the area were esti-
mated, agronomic trials were harvested, and the maize
recommendations were refined. The Third Annual Meeting
was held in early January 1970, attended by political
leaders, representatives of agricultural institutions, farmers,
and Project staff.

Project Operations: 1970

A principal addition to the Project in 1970 was the
initiation of technical assistance to farmers in Zone V (Fig.
1.2). A fifth technical assistance agent was assigned to this
The main thrust of the technical assistance program in
1970 was to increase the use of Project recommendations
by (a) informing farmers of the new technology through
village meetings and mass communications media, and (b)
assisting in the formation of new groups and in helping
organized farmers arrange for credit and fertilizers. A
special effort was made to increase the participation of the
public credit banks by counseling new groups to seek
financing from these sources, and by assisting the groups in
making the necessary arrangements.
The agronomic research program was expanded in 1970
to include studies on the production of beans and alterna-
tive crops for late plantings. The investigations of the
production of these crops were made in response to re-
quests from many of the farmer groups that had used the
maize recommendations the previous year.
In early 1970, CIMMYT signed an agreement with the
United Nations Development Program that led to the par-
ticipation of the Puebla Project and its advisors in the
development of similar programs in other areas (refer to
Chapter 13). The Project and its advisors participated

mainly in two activities: (a) the training of professional
staff for new programs in the philosophy, organization, and
operation of the Puebla Project (refer to Chapter 12); and
(b) technical assistance to the staff of new programs in
organization and agronomic research. A group of tech-
nicians from Colombia and Mexico were in training in Pue-
bla during the second half of 1970. Project advisors traveled
to Colombia and Peru on several occasions during the year
to assist in planning and organizing programs in those
Project staff and advisors participated in two interna-
tional conferences held in Puebla in August 1970, to discuss
strategies for increasing agricultural production on small
holdings. These conferences (English spoken at one,
Spanish at the other) drew participants from 15 Latin
American countries and 15 international development
Demonstrations at harvest time, estimations of yield,
and adjustment of maize recommendations were conducted
as in previous years. In addition, recommendations for the
production of bush beans were formulated for parts of the
Project area. The Fourth Annual Meeting was held in mid-
February 1971.

Project Operations: 1971

The operation of the Project in 1971 was similar to
1970, with one principal difference: a second personal
interview survey was conducted during the summer of 1971
to collect data for evaluating social and economic changes
in the Project area. Interviews were made of a sample of all
farmers in the area and a second sample of those farmers on
credit lists.
The technical assistance agents broadened their activities
in response to the requests from the farmers for assistance
in improving other agricultural activities. Short courses for
farmers on the management of orchards were given, with
the participation of specialists from other institutions.
Demonstrations were held to instruct farmers in the use of
small trench silos for preserving maize stover as silage.
Several groups were assisted in arranging for long-term loans
to finance the drilling of wells, or for purchase of farm
Technicians from Colombia, Ecuador, Peru, and Mexico
were trained in Puebla during 1971. Project staff and
advisors provided technical assistance to regional produc-
tion programs in Colombia, Peru, Honduras, and the States
of Mexico and Tlaxcala in Mexico.
The Fifth Annual Meeting was held in Puebla in mid-
February 1972.

Project Operations: 1972

In early 1972, the maize breeding component of the
Puebla Project was discontinued. Progress in developing
higher-yielding varieties had been slow (refer to Chapter 4).
Moreover, the maize breeders in CIMMYT felt that this
type of research could best be done at a research center

with ecological conditions similar to the Project area.
Varietal testing continued and became a part of the agron-
omic research program.
Results obtained in preliminary studies of the maize-pole
bean association in 1970 and 1971 indicated that net
income might be greater with the combination of crops
than with either maize or bush beans alone. Research was
intensified on management practices for this association, to
develop recommendations for its use.
The effort to achieve a closer coordination of the activi-
ties of the farmers, agricultural institutions, and Project
staff in previous years had disclosed problems that required
changes in the operating procedures of institutions. At
times, through a careful exposition of the problem to the
indicated institution, it had been possible to reach a sat-
isfactory solution. In other cases, however, little or no
progress had been made. It became clear from these experi-
ences that a more detailed study was needed of the operat-
ing procedures of the agricultural institutions and of the
reasons farmers have difficulty in using their services. Such
a study was undertaken in mid-1972.
Another increasing concern of the Project staff was how
to develop the capacity of the farmer organizations to
participate more actively in seeking solutions to their
production problems. In mid-1972, arrangements were
made to contract the services of a sociologist with years of
experience with farmer organizations, who would provide
technical assistance to the Project staff in searching for a
more effective way of working with the farmer groups.
CIMMYT decided in early 1972 to terminate its partic-
ipation in the Puebla Project at the end of 1973. The
Project had begun in 1967 as an experiment to learn how to
rapidly increase maize production among small, low-income
farmers. As the Project evolved, however, it became clear
that the Project's objectives would shift to more efficient
strategies for increasing production, net income, and the
general welfare of small farmers in rainfed areas. CIMMYT
felt that its mandate was not broad enough to encompass
all the activities that clearly should be incorporated in so
extensive an undertaking. This position was made known to
the Governor of Puebla and the Secretary of Agriculture,
making clear CIMMYT's reasons for withdrawing support,
as well as the conviction that the Project should continue.
A seminar on rural development was held in Bogota,
Colombia, in September 1972, with the participation of
representatives of the Puebla Project and similar programs
in Colombia, Honduras, Peru, and Mexico. The Interna-
tional Development Research Center (IDRC) of Canada
S took the initiative in organizing the seminar and provided
financial support.
The Sixth Annual Meeting was held in Puebla in March

Project Operations: 1973

The program of technical assistance to farmers was,
broadened in 1973 to include promotion of the use of: (a)
a new technology for the maize-pole bean association, and
(b) agua ammonia as a source of nitrogen for maize. The

national fertilizer agency, Guanos y Fertilizantes, intro-
duced the idea of finding a way to enable small, traditional
farmers to use agua ammonia. The agency provided the
agua ammonia, a source of phosphorus, applicators, credit,
and technical assistance. The Project technical assistance
agents informed the farmers of the availability of credit for
these fertilizers and assisted interested farmers in organizing
and in arranging for and applying the materials.
During the second half of 1973, the technical assistance
agents, after months of discussions about how to work more
effectively with farmer organizations, began to work more
intensively with selected groups. A series of meetings was
held in which the farmers and the technical assistance agent
explored operations that could increase net income of the
farmers. From these discussions, it could usually be agreed
that one activity should be given priority. A committee was
elected by the farmers to study how to proceed on the
priority undertaking. The technical assistance agent pro-
vided information and guidance to these groups, but re-
sponsibility for group action remained with the farmers.
A second seminar on rural development, with financial
support from IDRC, was held at Chapingo, Mexico, in
September 1973. The staff and advisors of the Puebla
Project participated in the organization of the seminar and
in the conferences and work sessions.
In early 1973, the Project staff and advisors began to
explore, with political leaders of the state and federal
governments, means for continuing the Puebla Project after
CIMMYT had terminated its participation. Discussions and
study at several levels proceeded during the year, and
shortly before the end of 1973, the Ministry of Agriculture
decided to continue the Project as an activity of the
Graduate College at Chapingo. Present expectations are that
the Project will become a part of a new national program to
increase agricultural production in rainfed areas of the


The coordinator has been the central figure in the
operation of the Puebla Project. His function has been that
of coordinating activities of the farmers, agricultural institu-
tions, and Project staff so as to enable small farmers to
attain higher levels of production and net income. In prac-
tice, his responsibilities have included three distinct (but
closely related) activities: (a) administration of the pro-
gram, (b) direction of the program, and (c) acquisition and
maintenance of the full support of the agricultural institu-
The Project coordinator made- most of the decisions
affecting the administration of the Project. He was responsi-
ble for locating candidates for staff positions, evaluating
their qualifications, and deciding whom should be hired. He
recommended salary levels and perquisites for Project staff,
the purchase of vehicles and equipment, budget changes,
etc. He approved the local expenditure of funds for day
laborers, supplies, gasoline, vehicle maintenance, etc.
The coordinator directed the activities of the Project
staff in the: (a) preparation of operational plans, (b)
execution of these plans, and (c) the summarizing and
reporting of accomplishments. Each program presented its
plans for the year ,at meetings of staff and advisors that
were generally held in January. The plans were discussed,
modified and finally approved. Weekly meetings were held
throughout the year to discuss progress and problems of the
staff. Adjustments in operational plans, as a result of new
information, were made at these meetings. Important
matters affecting the operation of the Project were discus-
sed fully at the weekly meetings before a decision was made
by the coordinator. The coordinator maintained contact
with the field work by accompanying the members of the
staff, as time permitted, in their daily activities. At the end
of the year, the coordinator worked with the staff members
in analyzing, evaluating, and reporting the results of their
A large part of the coordinator's time was dedicated to
work with the agricultural institutions. Initially he was
involved in informing the institutions of the philosophy,
objectives, and plans of the Project, and becoming familiar
with their operating procedures. Then, as information
flowed in from the field work, much of this data had to be
communicated to. the institutions. For 2 or 3 months after
harvest, the coordinator was in almost constant contact
with representatives of the institutions, explaining the plan
of operations for the following season and working to
obtain their approval and support. When a problem arose
due to the operating procedures of an institution, informa-
tion about the problem was prepared by the Project staff
and communicated by the coordinator to the responsible
people. Generally this was followed by a series or meetings
and the gathering of additional data until a decision could
be made.


From the beginning it was recognized.that the quality of
the Project staff would be the most important factor in
assuring the success of the undertaking. Screening proce-
dures were followed which, hopefully, would assure the
selection of the best candidates available.
The Project sought to provide working conditions and
opportunities that would enable its staff to work har-
moniously and effectively and to advance professionally
through (a) salaries and perquisites that were competitive
with other employment opportunities; (b) opportune avail-
ability of the necessities for getting the job done (adequate
operating expenses for vehicles, prompt purchase of
equipment and supplies, revolving funds for the purchase of
small items, prompt repair of vehicles and equipment, etc.);
(c) encouragement to use initiative and be innovative (the
staff could not be provided with an operations manual that
would cover every exigency that might arise; thus, the team
members were encouraged to work out their own solutions
when confronted with new problems and to seek advice
from other staff and advisors as soon thereafter as possible);
and (d) opportunities for advancement (outstanding team
members were given the opportunity to advance both in
salary and in professional position; also, staff members
interested in continuing their academic preparation were
assisted in doing so, after 2 to 3 years with the Project).
Because of the Project policy encouraging its staff mem-
bers to continue their academic training, plus the avail-
ability of other job opportunities, especially in regional
production programs in other parts of Mexico, there were
frequent changes. in Project personnel. Figure 2.1 shows
diagrammatically the periods of employment of profes-
sional staff during the period 1967-1973. The shortest
period of service was one cropping season; the longest
service was.6 years and 4 months.
When possible, new staff members were hired 1 to 3
months before the resignation of the person they would
replace. Thus, it was possible for the departing staff mem-
ber to relay to the replacement much of the knowledge that
had been gained of the area, farmers, and institutions. Also,
many departing staff members entered the Graduate Col-
lege at Chapingo and continued to be available for advice
and information.
The lines in Figure 2.1 show that there were 35 periods
of employment in the Project. However, one staff member
served both in evaluation and coordination, and a second in
both evaluation and technical assistance; thus, there was a
total of 33 staff members during the 7-year period. The
total number of man-years of professional time varied from
2.2 in 1967 to 12.3 in 1971.
Beginning in 1967, young farmers in the Project area
were hired to assist in the field activities. Initially they were
hired as day laborers, but some of them were given perma-
nent employment after a period of training and selection.

During 1967-1970, about 15 outstanding farmers were
selected as permanent field assistants. Other farmers were
hired as day laborers during the planting and harvesting
seasons. As an average for the 1967-1973 period, the
Project used approximately 25 man-years of the time of
these employees.


Specialists in agronomic research, maize breeding, and
communications, at CIMMYT and the Graduate College at
Chapingo, served as advisors to the Puebla Project. They
drafted the original Project proposal, arranged for funding,
selected the Project area, made the necessary arrangements
with local institutions, prepared operational plans for 1967,
and employed the first members of the Project team.
During the course of the Project, the advisors have as-
sisted the Project staff in: (a) evaluating and modifying
operational strategies, (b) preparing detailed plans for the

Project at the beginning of each year, (c) defining appropri-
ate methodological procedures, (d) resolving problems that
have obstructed progress, (e) analysing and interpreting
research findings, etc. In addition, the advisors have
complemented the activities of the Project staff by: (a)
anticipating Project growth and requesting funds needed for
expansion; (b) giving wide distribution to Project findings
through personal contacts, publications, and conferences;
(c) projecting the Puebla approach to other areas in
Mexico and Latin America by obtaining funds for travel,
fellowships, equipment, etc.; (d) informing agricultural and
political leaders of the Puebla experience; (e) planning and
participating in a training facility at Puebla; and (f) provid-
ing assistance in the organization and operation of new
The total time provided by the advisors in direct as-
sistance in organizing and operating the Puebla Project
averaged approximately 172 man-days per year for the
1967-1973 period.

Fig. 2.1. Periods of employment of professional staff in the five programs of the Puebla Project. Each line represents the
period of employment of a staff member.

PROGRAM 1967 1968 1969 1970 1971 1972 1973






TOTAL NUMBER OF MAN-YEARS 2.2 6.0 10.8 10.2 12.3 9.9 10.5

Field experiments were closely supervised during the crop
growing season, and observations on factors affecting yield
were made periodically. Here the height of unfertilized
maize plants is being measured. To the right is a plot show-
ing a strong nitrogen response.




Crop production on a given area depends on several
factors, including soil and climatic conditions, plant variety,
and production practices. The physical environment cannot
be changed readily and thus determines the yield potential
of a region. Varietal characteristics and management prac-
tices, however, are more easily manipulated, and improve-
ment of these factors can provide higher yields and net
income. Thus, agronomic research in the Puebla Project was
designed to provide information on soil management prac-
tices and varieties that would produce higher returns for the
farmers' production investments.


Production practices that can greatly influence crop
yields in rainfed areas include: land preparation; planting
date; seeding rate; amount and kind of fertilizers applied;
time and method of applying the fertilizers; control
measures for weeds, insects, rodents, and diseases; and
depth of plowing. As a first step in developing better
information on these agronomic practices, the Project
sought to gather as much information as possible on farm-
ers' production practices, soil and climatic characteristics,
and the experiences of other researchers in the area. This
information was gained by interviewing farmers and agron-
omists residing in the area, by reviewing the research find-
ings of the National Agricultural Research Institute, by
analyzing the available climatic data, and by studying the
properties of the soils in the area. Through this process an
understanding was obtained of the physical environment
and the traditional technology of the farmers, as presented
in Chapter 1.
A list of technological questions to be investigated was
prepared and arranged in order of priority by taking the
information available at the beginning of a given cropping
season and estimating: (a) probable deficiencies in existing
technology, (b) probable improvements that could be made
economically, (c) which improvements would most likely
provide the largest increases in yield and net income, and
(d) which of these aspects of the technology should-be
investigated in the Puebla area.
Next field experiments were carried out to answer ques-
tions of highest priority. The ecological diversity of the area
was taken into account in planning the research and in
locating the field trials. In 1967 and 1968, information on

the physical environment was limited, and field experi-
ments were distributed fairly evenly over the area. In 1969
and afterward, two or more producing systems were recog-
nized in the area, and the experiments were located to
sample these systems. (Note that a producing system is
defined as a part of a production universe in which the
uncontrollable production factors for a cropare reasonably
constant. These factors include: soil morphology, geomor-
phology, climate, previous crop, and at times, planting
During the growing season, observations were made
periodically on the factors influencing production at each
experimental site. The trials were harvested, data analyzed,
and results expressed as treatment means or production
functions. Data on crop response to rates of fertilization
and plant density were expressed both as mathematical
functions and as two-dimensional graphs, for greater cer-
*tainty and ease of interpretation.
In 1967, the information available from the National
Agricultural Research Institute was taken as a first approxi-
mation to the recommendations on crop production
practices for the Puebla area. (The recommendation for
rainfed maize in Puebla was: fertilize with 80 kg N and 40
kg P205/ha, use the hybrid H-28 with 40,000 plants/ha,
and plant at the beginning of the rainy season.) Data col-
lected in 1967 were used to revise the existing recom-
mendations on maize production and to calculate a second
approximation to the recommended practices. The data
collected in subsequent years were used to generate third,
fourth, fifth, etc., approximations.
Beginning in 1969, maize recommendations were for-
mulated for distinct producing systems. All available infor-
mation on climatic variability and prices for maize and
inputs was taken into account in estimating the optimal
levels of practices. Recommendations were made after
carefully weighing: (a) the precision of the available infor-
mation on the relationship between yield and the produc-
tion factors, (b) the marginal productivity of the factors in
question, and (c) the rislrinvolved in making recommenda-
tions that might not be appropriate for the farmer.
The remainder of this chapter describes the agronomic
research in the Puebla area in each year, 1967-1973. A final
section seeks to evaluate the adequacy of the maize tech-
nology and to estimate the potential benefits of the im-
proved technology, as compared with technology existing
in 1967. Appendix A provides a benefit: cost analysis of
the agronomic research program in the Project.


Information available at the beginning of 1967 indicated
that maize production was the major agricultural activity in
the Project area, accounting for the greater part of the avail-
able land, labor, and capital. Preliminary findings suggested
that maize yields could be increased substantially under
most farming conditions by: (a) increasing the rate of
fertilization with nitrogen and phosphorus, (b) using higher
plant densities, and (c) using better weed and insect control
measures. There was evidence that fertilization with potas-
sium, although a common practice, was not contributing to
higher yields.
Based on existing knowledge, the initial reasoning was
that optimal levels of fertilization with nitrogen and phos-
phorus would be greatly influenced by local conditions,
thus should receive priority in the field research program in
1967. It was also concluded that optimum levels of factors
such as time and method of applying fertilizers, plant den-
sity, genotype, and plant protection, which are generally
less affected by local variations in soils and climate, could
be estimated a priori from experiences gained in similar
Thus, two hypotheses tested initially were: (a) produc-
tion of rainfed maize in most of the area was being limited
by the rates of nitrogen and phosphorus commonly used by
farmers, and (b) fertilizing with potassium or zinc did not
increase maize yields. In designing the experiments to test
these hypotheses it was assumed that: (a) a population
density of 50,000 plants/ha would be near optimal for the
higher levels of fertilization with nitrogen and phosphorus
that were thought to be needed, (b) the maize plantings
should be kept free of weeds for the first 60 days after

emergence, and (c) high-yielding local varieties were avail-
able for use in the experiments.
A field experiment was designed and installed at 23
locations distributed throughout the Project area. Treat-
ments were used to measure the response of maize to levels
of nitrogen and phosphorus fertilizer. The criteria of the
cooperating farmers were used in deciding land preparation
practices, the date and method of planting, and times of
cultivating each experimental site.
The field trials were begun between April 18 and May
11. A composite soil sample was collected at each site. Rain
gauges were installed near each experiment and arrange-
ments made for the cooperating farmers to maintain a
record of daily rainfall. The experiments were visited
regularly during the growing season and data were collected
on: (a) dates of all farming operations; (b) phenological
dates; (c) vegetative response to treatments; and (d) damage
due to drought, hail, frosts, lodging, excess water, weed and
insect infestations, and diseases. A pit was dug at each
experimental site and a description was made of the soil
The experiments were harvested between October 6 and
25. The ears were harvested from the center rows of each
plot, weighed, and the moisture content of the grain was
determined. Observations were made on the percentage of
rotten kernels, the pollination percentage, and the shelling

Results: 1967

The data collected during the growing season on factors
influencing maize development were summarized for four
distinct parts of the growth cycle: (a) from planting to 45
days prior to flowering, (b) the 45 days preceding flower-
ing, (c) the 45 days after flowering, and (d) the period from
the 46th day after flowering to physiological maturity.


~rL~a. ~L+I

Field experiments were con-
ducted in cooperation with
farmers at sites carefully
selected to sample the varia-
tion in soils, climate and past
management existing in the
area. Here bags with different
fertilizer treatments are being
placed in the experimental
plots just prior to application.

As soon as the maize reached
maturity, the ears were har-
vested from the experimental
plots, weighed, and grain sam-
ples were taken to determine
the moisture content. The
yield data from the field ex-
periments were analyzed,
interpreted, and used to de-
velop new recommendations
on crop production practices.

The growth cycle was subdivided in this way because the
magnitude of the effect of most factors on maize yields has
been shown to depend on the stage of development of the
plant at the time the damage occurs. The effect of drought,
for example, is greatest when it occurs during the second or
third parts of the growing period. (Throughout this chapter,
data are presented for these four parts of the growth cycle
in their successive order, from planting through physiolog-
ical maturity.)
For the 23 experiments conducted in 1967, on the aver-
age there were 0, 7.3, 6.4, and 0.1 days with visible wilting
during the first, second, third, and fourth parts of the
growing period, respectively. The highest frequency of
drought occurred precisely in the two periods when maize
is most susceptible to damage. In general, there was little
damage due to hail, high winds, and frost.
Maize yields were increased significantly by the applica-
tion of nitrogen and phosphorus in 21 of the 23 experi-
ments. Fertilizers did not increase yields in one experiment
where the soil was naturally very fertile, nor at a second
location where drought was severe and a poorly adapted
variety was used.
Average treatment yields in each experiment were used
to calculate a quadratic equation with maize yield ex-
pressed as a function of rates of nitrogen and phosphorus.
These equations were used to estimate the optimal rates of
fertilizer for each experiment. The partial derivatives of
yield with respect to nitrogen and phosphorus were equated
to the ratio of the cost of the corresponding fertilizer to the
price of the maize. This resulted in two equations in two
unknowns whose simultaneous solution gave the optimal
rates of nitrogen and phosphorus for each experiment.
The estimated optimal rates of nitrogen in the 23 experi-
ments varied from 0 to 221 kg/ha, with an average of 109
kg/ha. Optimal rates of phosphorus varied from 0 to 128 kg
P2 05 /ha, with an average of 30 kg/ha. The maize yields


were calculated for each experiment corresponding to the
estimated optimal rates of nitrogen and phosphorus; these
varied between 2,128 and 7,068 kg/ha grain, with an aver-
age of 4,137 kg/ha. The average yield without fertilizer in
the 23 experiments was 1,326 kg/ha. Thus, the average
increase in yield produced by the estimated optimal levels
of fertilization was 2,811 kg/ha.
Results of the 1967 experiments were used to arrive at a
second approximation to the recommended package of
production practices for maize. Because a decision had been
made to limit promotional activities in 1968 and 1969 to
Zones I through IV, however, a second approximation was
derived specifically for that portion of the Project area (see
Figure 1.1). Fifteen of the experiments conducted in 1967
had been located in Zones I through IV. The optimal rates
of nitrogen for these 15 experiments varied from 60 to 221
kg/ha, with an average of 128 kg/ha. The optimal rates of
phosphorus for the same experiments varied from 0 to 128
kg/ha, P205, with an average of 37 kg/ha.
Two conditions suggested that the recommended levels
of nitrogen and phosphorus should probably be slightly
greater than the average optimal levels calculated from the
1967 results: (a) historical rainfall data and information
from farmers indicated that drought during the flowering
period of the maize crop (July and August) had been
unusually severe in 1967 it was probable that, in most
years, reductions in yield due to drought would be less than
those observed in 1967; and (b) one of the varieties used in
the experiments seemed poorly adapted in two locations-
responses to fertilization at those sites would probably have
been greater with a better-adapted variety.
It was decided, therefore, to recommend 130 kg/ha N
plus 40 kg/ha P20s for maize plantings in Zones I through
IV in 1968. One-tenth of the nitrogen and all the phos-
phorus were to be applied at planting time; the rest of the
nitrogen was to be applied just before the second cultiva-

tion. A population density of 50,000 plants/ha was to be
used, and the plantings were to be kept free of weeds
during the 60 days following emergence.
It was estimated that this revised recommendation
would produce an average increase in yield of 3,066 kg/ha.
Estimated costs of this package of practices, mainly fertiliz-
er costs, were equivalent in value to 1,795 kg/ha maize. The
expected average net increase in grain production was 1,271
kg/ha. Two additional sources of income would be associ-
ated with the use of the recommendation: (a) yields of
stover would be increased proportionately to that of grain
and could be sold or used on the farm; and (b) the higher
labor requirements for applying fertilizers, harvesting, and
shelling the maize would increase family employment and
family income.
The soil samples collected at the 23 experimental sites
were analyzed for nitrifiable nitrogen and available phos-
phorus. The experiments were divided into four groups,
depending on whether the levels of nitrifiable nitrogen and
available phosphorus were less than, or greater than, 10
parts per million parts of soil (ppm). The value of 10 ppm
was selected arbitrarily, to permit a comparison of soil test
levels and average optimal rates of nitrogen and phos-
phorus. The average optimal levels of nitrogen and phos-
phorus for the experiments in each group are shown in
Table 3.1.

TABLE 3.1. The average optimal fertilizer rates for soils
containing different amounts of nitrifiable nitrogen and
available phosphorus.
Available Weighted
phosphorus average
(ppm) N rate
(Bray P1 Method)

<10 >10

Nitrifiable < 10 141-49(10)* 130-9 (5) 137
(ppm) >10 90-37(2) 44-12(6) 55

average 47 10
P20 rate

* The first number is the average optimal rate of nitrogen, the second is
the average optimal rate of P205, and the third, in parenthesis, is the
number of experiments corresponding to the group.
The average optimal rates of nitrogen were 137 kg/ha for
soils containing less than 10 ppm of nitrifiable nitrogen and
55 kg/ha for soils containing more than 10 ppm. The aver-
age optimal rates of P205 were: 47 kg/ha for soils contain-
ing less than 10 ppm of available phosphorus; and 10 kg/ha
for soils containing more than 10 ppm. This promising
relationship between optimal rates of fertilization and levels
of available soil nutrients prompted the staff and consult-
ants of the Puebla Project to explore the possibility of using
soil analyses as an aid in determining fertilizer recommenda-
tions for Puebla farmers. Unfortunately, it was not possible
to provide an efficient soil testing service for the farmers,
and the Puebla Project was not able to make use of this


The experiments in 1967 suggested that under unfavor-
able conditions (severe drought, shallow soils), the popula-
tion density of 50,000 plants per hectare was probably too
high. For certain favorable production conditions (little or
no drought, deep soils) the same plant density appeared to
be too low. Thus, it was decided to study levels of plant
density along with levels of nitrogen and phosphorus.
It was also decided that experimental verification was
needed for the hypothesis that significant amounts of
moisture were conserved by fall plowing.
Observations of the traditional land preparation prac-
tices of the farmers during the winter of 1967-1968 led the
research staff to question the effectiveness of these prac-
tices for several reasons: (a) there is little weed growth
during the winter, thus little moisture should be lost, even
without plowing; (b) February and March are windy
months, and leaving the surface bare might foster wind
erosion; and (c) the organic matter contents of the soils are
very low, and plowing the soil would tend to accelerate the
mineralization of the organic matter.
Another question arose in 1967 about the way farmers
made their last cultivation. Most farmers cultivated very
deeply with a double moldboard plow and pruned many of
the lateral maize roots. This also seemed to be a factor for
local study.

Experiments on farmers' fields were used to obtain infor-
mation on rates of fertilization, time on applying nitrogen
and phosphorus, dates of planting, methods of land prepa-
ration, residual effects of fertilizers and manures, and other
production practices.

Deep, volcanic ash soils oc-
cupy about two-thirds of the
project area. By plowing in
the fall, farmers are able to
conserve much of the mois-
ture present in the soils at
harvest time. Just prior to
planting, the farmer plows
deep furrows and plants the
maize in holes opened with a
spade in the bottom of the

In parts of the Project area (particularly in Zone II),
much of the maize is grown in fruit orchards in the space
between rows of trees. It seemed likely that the effect of
the trees on the production of maize would vary depending
on: (a) the fruit specie, (b) the amount of space between
rows of trees, and (c) the size of the trees. Beginning in
1968, experiments were conducted to determine optimal
levels of fertilization for maize growing in fruit orchards,
taking into account the distinct characteristics of the
As shown in Table 3.2, plant densities, methods of land
preparation, depth of the last cultivation, and rates of ferti-
lization of maize in orchards were new lines of research in
1968. Studies on dates of planting were added when it
became evident in 1967 that farmers planted maize from
mid-March until late June. The research staff also decided
to include studies of times of applying fertilizers and the
residual effects of fertilizers to develop more reliable infor-
mation about these factors.
The research program in 1969 (Table 3.2) was similar to
that of 1968. The principal new line of research was the
study of application rates for manure and fertilizers. In
1968, it had been observed that farmers' plantings that had
received chicken manure that year, or chemical fertilizer
that year and chicken manure during the preceding three-
year period, were often more vigorous than the best experi-
mental treatments. This suggested the possibility of a
nutritional deficiency other than nitrogen and phosphorus.
It was decided to include experiments in 1969 to determine
economically optimal combinations of nitrogen, phos-
phorus, and chicken manure. In addition, the experiments
were planned so that residual effects of the manure could
be measured.
Two other lines of research were initiated in 1969: (a)
optimal rates of nitrogen, phosphorus, and plant density for
forage maize; and (b) effect of minimal tillage on maize
yields. In addition, the use of a "potential yield" treatment
in many of the experiments was begun that year. This treat-
ment consisted of 10 ton/ha of chicken manure, plus

t -rtA 't1'4 dC .-~

chemical fertilizers (140 or 160 kg/ha N plus 50 kg/ha
P205). It was assumed that this treatment would provide
all maize nutrition requirements.
A total of 47 field experiments were conducted in 1968
and 1969 (Table 3.2). Composite soil samples were collect-
ed at each experimental site from the plow layer (0-18 cm)
and from the subsoil (20-35 cm), for property characteriza-

TABLE 3.2. Lines of research in maize and numbers of
field experiments conducted in the Puebla area in 1968 and
Number of
Lines of research 1968 1969

Rates of nitrogen, phosphorus
and plant density 8 12
Rates of nitrogen, phosphorus and
plant density in maize for forage 0 1

Rates of nitrogen, phosphorus
and manure. 0 3

Dates of planting 4 2

Times of applying fertilizers 2 5

Depth of the last cultivation 2 0

Rates of nitrogen and phosphorus
for maize in orchards 2 2

Methods of land preparation 1 0

Minimal tillage 0 1

Residual effects of fertilizers 1 1

Total 20 27

Results: 1968 and 1969

Conditions in 1968 were favorable for the production of
maize. The average numbers of days per experiment with
plant wilting were 1.2, 1.7, 2.0, and 0 days during the four
successive parts of the growing cycle. In three of the 20
experimental sites there was slight damage due to hail
during vegetative development, but no damage during the
grain-filling period. There was slight or moderate frost
damage during the first part of the growing cycle in half of
the experiments.
In contrast, 1969 was a poor year for maize production.
Plant wilting occurred on an average of 14.1, 14.1, 0.5, and
0 days during the four parts of the growing cycle. In several
of the experiments, the plants began to wilt two weeks
after emergence and continued under moisture stress until a
week or so before tasseling. Slight to severe hail damage
occurred in one-third of the experiments during the first,
second, or third parts of the growing period. Frost did not
affect the maize plantings in 1969.
Studies of soil morphology done in 1968 revealed a large
region in the northwestern part of the Project area in which
the subsoils were sufficiently dense and compacted to re-
strict water movement and the penetration of maize roots.
Two producing systems were thus recognized and taken
into account in the interpretation of the experimental re-
sults in 1968: (a) deep soils of Popocatepetl, comprising
Zones III, IV, and parts of I and II; and (b) soils with a
compact layer impeding root development comprising parts
of Zones I and II.
The economically optimal rates of nitrogen for the
experiments conducted in deep soils in 1968 varied from
133-200 kg/ha, with an average of 187 kg/ha. The optimal
rates of phosphorus varied from 50-100 kg/ha P205, with
an average of 81 kg/ha. Optimal population densities varied
from 42,000-70,000 plants/ha with an average of 64,400
plants/ha. Grain yields, using these optimal treatments,
varied from 4,510-8,790 kg/ha, with an average of 7,462
kg/ha. The average increase in grain yield above the control
treatment was 6,434 kg/ha.
For the producing system in which the soils have a
compacted horizon, the average optimal rates of nitrogen,
phosphorus, and population density were 106 kg/ha, 58
kg/ha P20 and 55,333 plants/ha, respectively. The average
yield obtained using the optimal treatments was 4,847
kg/ha grain; the average increase over the control treatment
was 3,443 kg/ha. Clearly, for a favorable year like 1968, the
two producing systems differ markedly both in their poten-
tial for maize production and in the optimal rates of nitro-
gen fertilization.
The overall average grain yield in 1968 using the optimal
treatments was 5,312 kg/ha more than that obtained with
the control treatment. This compares with an average in-
crease of 3,292 kg/ha in 1967 for the experiments con-
ducted in the same area. This difference was due primarily
to more favorable rainfall in 1968 and the flexibility intro-
duced by including plant density in the experimental ma-

The results obtained in 1968 indicated that the recom-
mended rates of nitrogen, phosphorus, and plant density
should be increased for the producing system with deep
soils. However, because 1968 was a very favorable year, it
was decided to increase only the rate of phosphorus by 10
kg/ha P205.
For the system with soils having a compacted horizon,
the revised recommendation was for 20 kg/ha less nitrogen
and 10 kg/ha more P20s. Thus, the third approximation of
the maize recommendations was: (a) 130 kg/ha N, 50 kg/ha
P2Os, and 50,000 plants/ha for deep soils; and (b) 110
kg/ha N, 50 kg/ha P 2O and 50,000 plants/ha for soils
with a compacted horizon. For the other production prac-
tices, the recommendation remained the same as in the
previous year.
Studies of soil morphology done in 1969 identified a
large area of sodic-like soils in Zone IV. It was also found
that most of the soils in Zone V had formed on volcanic ash
derived from the volcano, La Malinche. According to field
response to fertilizers, these soils were well supplied with
phosphorus. Thus, four producing systems were recognized
in the interpretation of experimental results in 1969, in-
cluding: (a) deep soils of Popocat4petl; (b) soils of La
Malinche, and two other systems with soils having a com-
pacted layer impeding root development; (c) those compris-
ing parts of Zones I and II and recognized first in 1968; and
(d) the sodic-like soils in Zone IV.
Even with the unfavorable rainfall regime in 1969, the
deep soils of Popocat4petl yielded well and reflected the
need for a high rate of nitrogen fertilization. The best yields
obtained in 1969 on the soils with a compacted horizon in
Zones I and II were much lower than those obtained in
1968. The soils of La Malinche showed a lower yielding
potential than the deep soils of Popocat4petl under the
unfavorable climatic conditions in 1969. Even though
maize yields were increased by fertilization on the sodic-
like soils, maximum grain yields on these soils barely
reached 1,500 kg/ha. Maize yield on the sodic-like soils
were limited mainly by excess moisture during the first
three parts of the growing season.
The studies on rates of fertilization of maize in orchards
showed that the two rows of maize on either side of the
rows of trees should receive less fertilizer than the other
rows. The fourth approximation of the maize recommenda-
tions is presented in Table 3.3.
The study of the traditional method of land preparation
provided strong evidence that the moisture content of the
soil and the per cent emergence of maize planted the fol-
lowing spring were lower when the soil was not plowed
during the late fall, than with traditional practices. The
study of depth of plowing detected no significant reduction
in yield due to deep plowing at the last cultivation.
Date of planting and time of applying nitrogen showed
contrasting effects on yield in 1968 and 1969. Maize plant-
ed during the first week of April in 1968 yielded 2,000
kg/ha more than maize planted three weeks later. In 1969,
maize planted on the later date yielded 1,000 kg/ha more
than that planted three weeks earlier. In 1968, maize yields
were 600 kg/ha higher when 150 kg/ha N were applied at

the second cultivation, rather than at planting time. In
1969, yields were 1,200 kg/ha higher when 150 kg/ha N
were applied at planting time, rather than at the second
These contrasting effects of date of planting and time of
applying nitrogen seem to stem from differences in the
amount and distribution of rainfall during the two years.
The monthly rainfall for 1968 and 1969, and the monthly
averages for the 1941-1968 period are shown graphically in
Figure 3.1.
In 1968, rainfall in May was average, and in June it was
20 percent above the average. Thus, adequate moisture was
available in 1968 during the vegetative development of
early plantings of maize. In 1969, on the other hand, rain-
fall in May was 40 percent of the 1941-1968 average; in
June, 27 per cent; and in July, 73 per cent of that average.
Thus, early plantings of maize suffered severe moisture
stress during May, June, and early July.
The higher yields from the later plantings in 1969 appear
to be due to the fact that these plantings had a longer
period in which to produce a large plant than did the early
plantings, after the rains began in July. The better response
to nitrogen applied at planting time in 1969 was probably
because moisture deficiencies delayed absorption of nitro-
gen applied at the second cultivation until the plants were
too old to make maximum use of the nitrogen.
In 1969, it also was noted that certain maize varieties,
after being under moisture stress for several weeks, were
able to delay tasseling for one to two weeks, thereby devel-
oping larger plants and producing higher yields. This char-
acteristic of "latency" was important in 1969, but not in

---- 1968
....... 1969
-- 1941-1968

/ \
C /
S200 /
Ko /0 ~


Fig. 3.1. Average monthly rainfall in Zones I-IV of the Pue-
bla Project for 1968, 1969, and the period 1941-1968.


The interactions observed between rainfall pattern and
date of planting, time of applying nitrogen, and maize varie-
ty, in 1968 and 1969 suggested that it would be advanta-
geous to make integrated studies of yield response to these
variables, plus rate of fertilization and plant density. Six
such integrated studies were conducted in 1970, along with
studies of maize response to rates of nitrogen, phosphorus,
and plant density in six experiments located to sample
regions that had not been studied previously.
Research on bean production was initiated in 1970.
Major emphasis was placed on bush beans (determinate
growth habit) and lesser attention was given to the associa-
tion of maize with pole beans (indeterminate growth habit).

TABLE 3.3. Fourth approximation of the maize recommendations for the Puebla area.

Kg/ha of fertilizers applied at: Total
Planting First Second applied Population
time cultivation cultivation (kg/ha) density
Producing system N P2 O N P20O N POs N P2Os (Plants/ha)

1. Deep soils of Popocat6petl
1.1. Maizealone 30 50 0 0 100 0 130 50 50,000
1.2. Maize in orchards
1.2.1. Two rows
on either side
of the trees 30 50 0 0 50 0 80 50 50,000
1.2.2. Other rows 30 50 0 0 100 0 130 50 50,000
2. Soils of La Malinche 0 0 80 0 0 0 80 0 40,000
3. Soils with a compacted
layer impeding root
3.1. Non-sodic 20 50 0 0 90 0 110 50 50,000
3.2. Sodic-like Do not grow maize
Do not grow maize


A collection was made of 24 local bush varieties and 18
pole varieties. The response of bush beans to rates of fertili-
zation and plant density was studied at six locations. The
varieties collected in Puebla were evaluated at three sites.
The maize-bean association was studied at one location.

Results: 1970

The rainfall pattern in 1970 was quite favorable for both
maize and beans. In the several experiments, maize wilted
an average of 6.2, 0.7, 0, and 0 days during the four parts
of the growing cycle. Hail damage occurred in one-third of
the experiments during the first part of the growing season;
in the other three parts, hail affected the maize in only
one-tenth of the experiments. In most of the plantings,
maize was damaged slightly by frost during the first part of
the growing period.
Table 3.4 shows the average maize yields obtained at
two locations using several combinations of fertilization,
plant density, genotype, and date of planting. The average
grain yields for the 16 treatments were 5,352 kg/ha for the
first planting date, 4,446 kg/ha for the second date, and
2,029 kg/ha for the third.
There were large interactions between planting date and
other factors, including: (a) Planting Date x Rates of Fer-
tilization and Plant Density: comparison of the yields
obtained with Treatments 2 and 14 shows that the higher
rate of fertilization and plant density outyielded the lower
rate by 3,703 kg/ha in the first planting, 3,166 kg/ha in the

TABLE 3.4. Average maize yields obtained at two locations
genotype, and date of planting.

second planting, and by only 575 kg/ha in the third plant-
ing. A similar comparison of the Control Treatment (1)
and the "Potential Yield" Treatment (15) indicates that
Treatment I yields (no fertilizer) were 15 per cent of Treat-
ment 15 yields for the first planting date, 12 per cent for
the second date; and 6 per cent for the third. This seems to
imply that a given soil can supply more nutrients to early
plantings than to later plantings; (b) Planting Date x Fertili-
zation with Phosphorus: the magnitude of this interaction
can be judged by comparing yields of Treatments 4 and 5,
Treatments 7 and 9, and Treatments 8 and 10, at the three
dates of planting. The increase in yield due to phosphorus is
less than the least significant difference (LSD) for the first
planting date, and is clearly greater than the least significant
difference for the second and third planting dates. In these
experiments the same soil required less phosphorus and
produced higher yields in early plantings than in later plant-
ings; (c) Planting Date x Genotype: comparison of treat-
ments 15 and 16 shows that the yield obtained with the
Composite A x B was 91 percent of that produced by the
hybrid maize for the first planting date, 120 per cent for
the second date, and 135 per cent for the third; and (d)
Planting Date x Kind of Fertilizer: comparisons of treat-
ments 14 and 15 for the first and second plantings, and
Treatments 12 and 15 for the third planting, show that the
best yield obtained with chemical fertilizers alone was 82
per cent of that obtained with chemical fertilizers plus
manure for the first planting date, 76 per cent for the
second date, and 54 per cent for the third.

using several combinations of fertilization, plant density,

Yield in kg/ha
Nitrogen (kg/ha) applied at: Phosphorus Population Planting date
No. of Planting First Second (kg P205/ha) density Ma 23 June 13
Treatment 'time cultivation cultivation at planting (plants/ha) May 3 May 23 June 13

1 0 0 0 0 30,000 1306 932 234
2 0 50 0 25 30,000 3315 2771 1469
3 0 80 0 0 30,000 3869 2264 875
4 0 105 0 0 30,000 4303 2749 1128
5 0 105 0 40 30,000 4253 3788 1912
6 0 105 0 40 50,000 5399 4103 2235
7 0 130 0 0 30,000 4592 2641 1240
8 0 130 0 0 50,000 5539 3511 1600
9 0 130 0 40 30,000 5006 4048 1794
10 0 130 0 40 50,000 5907 4788 2358
11 30 0 100 40 50,000 6079 5311 1875
12 30 0 100 40 60,000 5989 4958 2179
13 30 0 130 40 60,000 6748 5946 2089
14 30 0 160 40 60,000 7018 5973 2044
15 30 0 130 40 60,000 8533 7872 4018
16** 30 0 130 40 60,000 7778 9475 5420




532 794 364

* The hybrid, H-129, was used for the May 3 planting. The hybrid, H-28, was used for the plantings on May 23 and June
** These treatments include an application of 10 ton/ha of chicken manure. The composite variety A x B (see Chapter 4,
page 40) was used in treatment 16.

Results similar to those shown in Table- 3.4 were ob-
tained in the other integrated studies. Based on.these data,
it was decided to make a separate recommendation for late
plantings. This recommendation would use 30 kg less nitro-
gen per hectare and 10,000 fewer plants per hectare than
the recommendation for early plantings. For soils with a
compacted horizon,'a small reduction in 'the phosphorus
rate wasalso recommended for late plantings.
A fifth approximation of the recommended production:
practices for maize was calculated" at the end of 1970,
taking into account all- information available-at that time.
Distinct packages of practices consisting of rates of fertiliza-
tion and plant density were recommended for 16 producing
systems, varying mainly because of differences in soil
morphology, planting date, and use for fruit trees.
Bush bean yields in the six experiments were influenced
by rates of fertilization and plant density. Average optimal
levels were 67 kg/ha N, 53 kg/ha P205, and 112,500
plants/ha. The average yield obtained with the optimal treat-
ments was 1,951 kg/ha beans. When no fertilizer was used,
the average yield was 780 kg/ha. The cost of the average
optimal treatment was equivalent to 605 kg beans with a
price of $0.12/kg and to 363 kg beans with a price of
$0.20/kg. Even at the lower price for beans, which rarely
occurs, the average increase in yield using the optimal treat-
ments was almost double the cost of the treatments.
These data were used to arrive at a first approximation
of production practices for bush beans: (a) for deep soils
of Popocatdpetl 'and soils with a compacted horizon: 60
kg/ha N, 60 kg/ha P205, and 120,000 plants per hectare;
and (b) for the soils of La Malinche: 60 kg/ha N, 30 kg/ha
P205, and 120,000 plants per hectare. The recommenda-
tions of the National Agricultural Research Institute for the
control of the bean beetle, Epilachnia varivestis (known
locally as "conchuela") were to be followed, with the farm-
er to select the variety, the date of planting, and the time to
Results obtained in the study of the maize-bean associa-
tion indicated that this cropping system might provide
greater net income to Puebla farmers than either maize or
beans grown alone.


Integrated studies of the effects of planting date, fertili-
zation, and plant density were continued at four locations
in 1971. It was decided to continue these experiments for
several years to accumulate information on the interaction
between these factors and climatic conditions.
Data from the integrated studies conducted in 1970
indicated that investigations of efficient management prac-
tices for late maize planting should take these factors into
account: (a) existing short-season varieties have a rela-
tively low-yielding ability, as compared to long-season
varieties; (b) there are probably nutrient deficiencies other
than nitrogen and phosphorus; and (c) light intensities and
temperatures are relatively low, and available soil moisture
abundant, in the initial stages of plant growth.

Three experiments were conducted at a single location in
1971 to determine optimal production practices for late
plantings of maize. These experiments covered three top-
ics; (a) exploration of the yielding ability of six varieties,
(b) study of, the response of a local maize variety to five'
Sminor elmehts, and (c) study of the response of an intro-
'duced maize vAriety to fertilization with nitrogen, phos-
:phorus,,andchickenn mahure,.and to plant density.
SThe response:of nae to five rates of nitrogen and phos-
phortis or nitrogen and plant: density'was studied at 10
locations to produce data for calculating the most adequate
mathematical model to represent maize response to these
factors (thesis research of a graduate student at Chapingo).
Experiments at six locations studied optimal levels of
fertilization and plant density for bush beans. Two experi-
ments compared net income from the maize-bean associa-
tion with that obtained from maize and beans grown alone
(pole beans were used in one experiment and bush beans in
the other).
A series of experiments at two locations sought to
identify crops that might be grown instead of maize in
years when the rains do not begin until July. Maize planted
as late as July runs a high risk of being damaged by frost in
the fall. Included in this series were sunflowers for forage,
bush beans, horse beans, oats, barley, and maize.

Results: 1971

The rainfall pattern in 1971 was quite favorable for both
maize and beans. In the maize experiments, the average
numbers of days with plant wilting were 8.7, 0.4, 0, and 0
for the four parts of the growing cycle. Slight to severe hail
damage occurred in the first, second, or third parts of the
growing cycle in 10 maize experiments. Slight to severe
frost damage occurred in the first part of the growing cycle
in four maize plantings. None of the bean experiments was
damaged, by drought, hail or frost. Bush beans suffered
moderate leaf damage due to anthracnose disease at three
Only one of the early-maturing maize varieties studied in
1971, Rojo Salvatori, showed a reasonably high yielding
ability. Table 3.5 compares this variety with Zacatecas 58,
which had the next highest yields. The Rojo Salvatori yield
at the highest level of fertilization and plant density was

TABLE 3.5. Grain yields obtained with two early-matur-
ing maize varieties receiving different fertilization and plant
density treatments.
Population Chicken
Nitrogen P205 density manure Grain yields (kg/ha) using:
kg/ha kg/ha plants/ha ton/ha Zacatecas 58 Rojo Salvatori

60 50 40,000 0 1237 1448
100 50 60,000 0 833 1840
150 80 80,000 0 1168 2870
120 80 80,000 10 2030 2537
200 100 100,000 0 1098 1676
200 100 120,000 0 1563 1354
150 100 120,000 20 1491 3147
150 100 150,000 20 2597 4317

three times that obtained at the lowest level of fertilization
and plant density.
The studies of the maize-bean association demonstrated
that net income from the association was approximately
double that obtained with either maize or beans alone.
Horse beans, bush beans, oats, and barley all showed some
advantage over maize for late plantings. The data obtained
in the study of the response of an early maize variety to
minor elements was inconclusive.
In 1971, the production of maize became somewhat
more profitable for farmers, due to several factors: (a) the
price of nitrogen fertilizers was reduced by about 14 per-
cent; (b) the maximum moisture content acceptable in
grain purchased at the guaranteed price was increased from
12 to 14 percent; (c) the practice of paying less for colored
grains was discontinued, and (d) the CONASUPO, the
National Marketing Agency, agreed to purchase maize in
small lots.
These changes prompted Project decisions to increase
the recommended rate of nitrogen fertilization in Zone V
from 80 to 100 kg/ha and the population density from
40,00b to 50,000 plants/ha. It was decided not to change
the recommendations for the rest of the Project area, but to
develop alternative recommendations that would cost about
one-half to two-thirds as much as the existing recommenda-

TABLE 3.6. Average yields, protein percentages, and net
experiments conducted at three locations in 1972.

tions. It was expected that the technical assistance agents
would provide information to the farmers about costs,
expected net incomes, and risks involved in the alternative
recommendations. The farmer would decide which alterna-
tive to adopt, or how much land to allot to each recom-
mendation. This innovation was designed initially for early
plantings (those made with residual moisture).


Results obtained with the maize-bean association in
1970 and 1971 were promising; thus, more resources were
allotted to the study of this cropping system. Six experi-
ments in 1972 measured the response of the association to
several rates of nitrogen, phosphorus, and plant density of
maize. These experiments were located in the important
producing systems of Zones I, II, and IV, where this
cropping system is commonly used.
Beginning in 1968, average maize yields were estimated
each year at harvest time on samples of two categories of
farmers: (a) farmers on credit lists (who were organized in
groups, received credit from institutions participating in the
Project, and could be expected to use the Project recom-
mendations), and (b) all farmers in the area (see Chapter 8).

incomes for several" treatments in maize-bean association

Population Grain with Net income with
density 14% moisture bean prices/ton at :
Treatment Fertilizers (kg/ha) applied: of maize* Percent
No. N P205 plants/ha Maize Beans protein** $240 $160

1 120 40 30,000 2987 1300 8.6; 22.0 404.64 300.64
2 120 40 40,000 3306 1246 410.88 311.20
3 120 80 30,000 2962 1548 440.72 316.80
4 120 80 40,000 3074 1393 8.3; 21.8 414.40 304.08
5 150 40 30,000 2796 1575 442.00 315.92
6 150 40 40,000 3758 1361 451.36 298.80
7 150 80 30,000 3006 1575 8.7; 22.4 441.44 317.84
8 150 80 40,000 3559 1398 8.7; 22.4 436.00 324.16
9 90 40 30,000 2619 1150 8.3; 21.0 357.60 265.60
10 180 80 40,000 3737 1488 9.0; 23.0 449.04 330.00
11 150 0 40,000 3156 1445 8.5; 22.9 440.48 324.88
12 150 80 20,000 2217 1641 397.92 266.64
13 150 40 +CM"+ 40,000 4056 2446 9.4; 24.5 557.04 361.36
14 120 40 40,000 4634 0 8.5 244.24 244.24
15 60 60 0 0 1222 20.9 194.08 96.40

LSD 575 225

Population of beans was constant at 60,000 plants per hectare.
** The protein percentage of maize appears first followed by that of beans. Each value is an average of 15 determinations.
The analyses were made by biochemist Francisco J. Rodriguez B. of the CIMMYT Protein Quality Laboratory.
+ Net income was calculated as gross income minus variable costs. The value of maize grain was calculated at $72.00 per ton
and the value of stover at $8.00 per ton.
+ Ten tons per hectare of chicken manure.

In general, average yields of farmers on credit lists were
only about two-thirds as large as they might have been,
according to the results obtained in the field experiments.
This finding suggested studies to determine why the farmers
on credit lists did not have higher yields.
In each of the Zones II and V, sixty parcels representing
60 farmers on credit lists were chosen at random. A repre-
sentative area was selected within each parcel consisting of
12 rows, 10 meters long. One of the two alternative maize
recommendations was used on six rows of each parcel in
Zone II, with the other alternative used on the other six
rows. In Zone V, the more costly alternative was used on
six rows of each parcel, with the same recommendation,
plus 50 kg/ha P205, used on the other six rows. The latter
treatment was included to test phosphorus needs of maize
fields in Zone V.
These two plots in each of the selected parcels were
managed by Project research agronomists. In addition, the
agronomists made regular observations of the production
practices used by the owners of the parcels, supplementing
this information with data collected directly from the farm-
Additionally in 1972, two field experiments were made
to: (a) determine if weed problems were greater in fields
using Project recommendations than in fields using the
traditional technology, and (b) evaluate the profitability of
the intensive weed control methods in the Project recom-
mendations. It had been found that many farmers had the
impression that weed control was more difficult in fields
where the Project recommendations were used, and there
was concern that this feeling might discourage farmers from
adopting the new technology.
The weed control experiments consisted of 12 treat-
ments. Project recommendations were used in half the
treatments and the traditional technology in the other half.
Several weed control measures were used with each of the
technologies. The more intensive weed control practices
'consisted of hand weeding at different growth stages, and
the use of herbicides. The experimental plot consisted of
six rows, each 5 meters long. A randomized complete block
design with six replications was used.

Results: 1972

Conditions in 1972 were excellent for maize and beans,
perhaps comparable to 1968. In 24 experimental plantings
of maize, or of maize associated with beans, the average
numbers of days with wilting of the maize plants were 0.5,
1.0, 0, and 0.9 for the four parts of the growing cycle. The
corresponding averages for 1968 were similar: 1.2, 1.7, 2.0,
and 0. Hail and frost damage in 1972 were also slight.
Table 3.6 shows average yields, protein percentages of
the grain, and net incomes for treatments used in three
maize-bean association experiments conducted in the deep
soils of Popocat4petl. The inclusion of 60,000 plants/ha of
beans in a planting of maize with 40,000 plants/ha, fertil-
ized with 120 kg/ha N and 40 kg/ha P205 resulted in: (a) a
decrease in the maize yield of 1,328 kg/ha (Treatment 14
minus Treatment 2); (b) production of 1,246 kg/ha beans;

and (c) an increase in net income per hectare of $166.64
with beans priced at $240/ton, or $66.96 with beans priced
at $160/ton.
Bean production increased remarkably when 10 ton/ha
of chicken manure was added to the treatment consisting
of 150 kg/ha N, 40 kg/ha P2C5, 40,000 plants/ha of maize,
and 60,000 plants/ha of beans. Comparison of bean yields
obtained with Treatments 6 and 13 shows that the increase
due to manure was 1,085 kg/ha. Part of this increase in
bean yield was probably due to the nitrogen and phos-
phorus contained in the manure. Most of the increase,
however, was apparently due to some different, undeter-
mined cause. The net income from the maize-bean associa-
tion receiving both chicken manure and chemical fertilizers
(Treatment 13) was 2.28 times that derived from maize
planted alone, with beans priced at $240/ton. The protein
percentage of the beans fertilized with chicken manure was
1.5 percent points higher than that of the beans receiving
the highest rate of chemical fertilizers (24.5 versus 23.0
Table 3.7 compares the amounts of protein and lysine
produced by common maize planted alone, and by the
maize-bean association, with the amounts produced by
opaque maize. Opaque maize produced nearly twice as
much lysine per hectare as did common maize. The com-
mon maize-bean association, fertilized with nitrogen and
phosphorus, produced 59 percent more lysine than did
opaque maize alone. The common maize-bean association,
fertilized with nitrogen, phosphorus, and chicken manure,
produced 2.39 times as much lysine as opaque maize alone.

TABLE 3.7. The amounts of protein and lysine produced
by common and opaque maize planted alone, and by the
common maize-bean association.
Protein Lysine *
%of % of
kg/ha Opaque kg/ha Opaque

Common maize, planted alone 394 93 9.9 52
Opaque maize, planted alone** 423 100 19.9 100
Maize-bean association with 150
kg/ha of N, 80 kg/ha of P205
and 40,000 plants/ha of maize 623 147 30.3 159

Maize-bean association with
150 kg/ha of N,
40 kg/ha of P205,
10 ton/ha of chicken manure
and 40,000 plants/ha of maize 981 232 45.5 239

* In calculating the production of lysine per hectare, the protein of
beans and common maize was assumed to have 7.2% and 2.5%lysine,
respectively. For bean protein from the treatment with the chicken
manure, it was assumed that the percentage of lysine dropped to 6%.
Source: Mercedes Hernandez, et ul., 1971. Valor nutritivo de los ali-
mentos; Tablas de uso practice. Publicaciones de la Divisi6n de Nutri-
ci6n-L-12, 5a. Institute Nacional de la Nutrici6n, M6xico. p. 20.
** Based on data from an experiment carried out on the deep soils of
Popocat4petl. The best opaque variety yielded 4,700 kg/ha with
50,000 plants/ha and fertilized with 130 kg/ha of nitrogen plus 50
kg/ha of P205. It was assumed that the grain contained 9% protein and
that the protein had 4.5%lysine.

The opaque maize would have had to yield 11.3.ton/ha to
equal the production of lysine by this maize-bean associa-
tion in 1972. An association of pole beans with an opaque
maize would seem certain to yield an even higher produc-
tion of lysine per hectare.
The study of the use of technology by the farmers on
credit lists disclosed that farmers' yields on the average
were lower than those obtained in the parcels controlled by
the research agronomists. The average yield of the farmers'
parcels in Zone II was 3,444 kg/ha and for the control plots
within these parcels it was 4,725 kg/ha. In Zone V, the
farmers' yields averaged 4,076 kg/ha versus a control yield
4,841 kg/ha. The lower yields obtained by farmers can be
attributed to failure to make full use of the recommended
technology. Only 28.1 percent of the farmers studied in
Zone II used at least three-quarters of the recommended
amounts of all three main inputs: nitrogen, phosphorus,
and plants per hectare. Another 19.3 percent of the Zone II
farmers applied at least three-quarters of the recommended

rates of fertilizer, but used fewer plants than recommended.
In Zone V, 56.9 percent of the farmers in the study used at
least three-quarters of the recommended amounts of nitro-
gen, phosphorus and plant density, with an additional
37.2 percent falling short only in the use of the recom-
mended plant density.
The data obtained in the weed-control experiments
refuted the hypothesis that the maize technology recom-
mended by the Project results in greater weed infestations
than traditional technology. When the more intensive weed
control measures of the new technology were used, the
weed population at' harvest time was lower than that ob-
tained with traditional methods, and increases in maize
production were proportionately greater.
A sixth approximation of the recommended production
practices was developed in early 1973. The recommenda-
tions for several producing systems in Zone II are shown in
Table 3.8, illustrating the type of information on produc-
tion practices available to farmers in 1973.

TABLE 3.8. The seventh approximation of the recommended production practices for several producing systems in Zone II.

Producing system Fertilizer (kg/ha) to apply at:
Maize Bean Variety
First Second population population of beans
Planting Level of Planting cultivation cultivation density Maize density or other
Soil morphology date Crop capital* N P205 N N plants/ha variety plants/ha crops

1. Deep soils of Po- Apr. 1- Maize I 30 40 0 100 50,000 H-131
pocat6petl. May 15 II 0-- 80 0 40,000 H-131
1.2. Free of pumice Maize in or-
on the surface; chards:
less than 2350 m a) 2 rows on
altitude, each side
of trees I 30-40 0 50 40,000 H-131
b) Other rows I 30-40 0 100 50,000 H-131

Apr. 25- Maize-bean I 30-40 0 120 40,000 native 60,000 native
May 15 association II 30- 0 0 60 30,000 native 45,000 native

Bunch beans I 60 60 0 0 -. 120,000 native

Mayl6- Maize and
June 15 maize in or-
chards I 30 -40 70 0 40,000 native

May 16- Maize-bean I 30 -40 120 0 40,000 native 60,000 native
June 15 association II 30- 30 60 0 30,000 native 45,000 native

May 16- Bunch beans I 60-60 0 0 -. -. 120,000 native
June 30

July 1- Bunch beans I 30 30 0 0 -. 90,000 native
July 15

June 16- Oats I 40-40 Use 90 kg of seed per hectare Cuauhtemoc
July 15 Barley I 30 30 Use 60 kg of seed per hectare Apizaco
Horse beans I 40 40 Plant density of 60,000 pts/ha INIA 15001

Recommendation I presupposes the availability to the farmer of unlimited capital for maize production; recommendation II presupposes limited capital of one-half to
two-thirds that required for the more costly recommendation.


This section seeks to evaluate the results of agronomic
research in terms of impact on maize yields, on net incomes
of farmers, and on the risks farmers take with input
investments for maize production. A major difficulty in
making such an evaluation stems from the fact that in-
creases in production and net income are generated by
many interacting factors (production credit, distribution of
inputs, markets, input cost: product price ratios, etc.), and
not by improved technology alone. Nevertheless, it seems
reasonable to examine the influence of project recom-
mendations by comparing increases in maize yields and net
income -and changes in risk as well -- that can be expected
if farmers adopt each of several production technologies
presently available.
According to the 1967 survey, 69.3 percent of the farm-
ers applied fertilizers to their maize plantings that year. The
production technology of those farmers (on the average)
consisted of approximately 50 kg/ha N, 25 kg/ha P205, 10
kg/ha K20 (potassium), 25,000 plants/ha, a local variety
and a planting date ranging from 0 to 75 days before the
beginning of the rainy season. Each of these production
factors showed variation across the Project area, probably
in response to the diversity of local producing conditions.
Unfortunately, the 1967 survey was not designed to collect
such information on the local production technology. Thus,
the average production technology of farmers is used here
to represent the traditional technology, allowing flexibility
only for the local variety and the planting date. The inflex-
ibility of the fertilizer treatment and the population density
assumptions is very probably biased against the traditional
technology in these comparisons. However, the assumption
that all farmers fertilized their maize should be a bias
favoring the traditional technology, since only 69.3 percent
of farmers applied fertilizer to their maize plantings in
A maize technology, recommended by the National
Agricultural Research Institute (INIA), was available to the
Puebla farmers in 1967. It consisted of 80 kg/ha N, 40
kg/ha P205, 40,000 plants/ha, the hybrid H-28, and plant-
ing as soon as the rains began. This recommendation
applied to the entire Project area.
The INIA technology was modified in two ways to
facilitate the comparison of technologies: (a) the recom-
mendation to plant at the beginning of the rainy season was
changed to "plant early (late March, April, early May) in
producing systems with adequate residual moisture." (It is
known, a posteriori, that early plantings of maize produce
higher yields. It was assumed that, once active promotion

of the INIA technology was underway, the Project staff
would recognize this fact and quickly change the recom-
mended date of planting). (b) the recommendation to plant
H-28 was discarded. This change was made because native
varieties (not H-28) were planted in most of the experi-
ments whose results were used to compare the different
technologies. It was thought that this change would intro-
duce little bias against the INIA technology, since local
varieties compare favorably with H-28 in their yielding
ability (see Chapter 4).
As indicated previously, the results obtained in the
experiments conducted in 1967 were used to develop a new
recommendation for maize, referred to here as the second
approximation to the maize production technology. Experi-
mental results obtained in subsequent years were used to
develop a third, fourth, fifth, and sixth approximation. The
sixth approximation, available at the beginning of 1972,
included recommendations for 16 maize-producing systems.
As shown in Table 3.9, these systems differed in soil
morphology, previous crop, elevation above sea level, or
planting date. Alternative recommendations for two levels
of capital investment were available for each of the 16
systems. The recommendation for the lower level of capital,
referred to here as the limited capital recommendation, was
selected rather intuitively; however, it corresponds closely
to the factor combination that maximizes the rate of
return on capital.
The 16 pairs of recommendations, together with their
variable costs expressed in tons of maize grain per hectare,

TABLE 3.9. The 16 producing systems recognized in the
Project area since 1972.

1.1.1 Deep soils of Popocat6petl; elevations between 2,100 and 2,350
meters above sea level; plantings before May 15.
1.1.2 Deep soils of Popocat6petl; elevations between 2,100 and 2,350
meters above sea level; plantings between May 16 and June 15.
1.2 Deep soils of Popocatepetl; elevations between 2,351 and 2,800
meters above sea level; plantings before April 30.
2.1.1 Pumiceous soils of PopocatBpetl; elevations between 2,100 and
2,350 meters above sea level; plantings before May 15.
2.1.2 Pumiceous soils of Popocatepetl; elevations between 2,100 and
2,350 meters above sea level; plantings between May 16 and
June 15.
2.2 Pumiceous soils of Popocat6petl; elevations between 2,351 and
2,800 meters above sea level; plantings before April 30.
3 Soils of La Malinche; plantings before April 5.
4 Heavy soils of Zone V; plantings at the start of the rainy season.
5.1.1 Soils with a compacted horizon; plantings made in March and April.
5.1.2 Soils with a compacted horizon; plantings made in May.
5.1.3 Soils with a compacted horizon; plantings made in June.
6.1.1 Sodic-like soils; plantings made in March and April.
6.1.2 Sodic-like soils; plantings made in May.
6.1.3 Sodic-like soils; plantings made in June.
7.1 Soils with a high Water table; plantings immediately after the
turning of alfalfa stubble; plantings before April 15.
7.2 Soils with a high water table or any irrigated soil; one or more
years after the turning of alfalfa stubble; plantings before April

are shown in Table 3.10. Variable costs were calculated on
the assumption that the farming operation was a mixed
enterprise. Hired labor was included as a cost, but labor by
family members was not included. For the calculations in
Table 3.10, it was assumed that hired labor included: one-
third of the labor at planting and the first fertilization;
one-half of the labor at the second cultivation and second
fertilization; three-fourths of the labor at harvest time; and
one-third of the labor for shelling. The expense of this labor
was included as a variable cost.
Several aspects of the unlimited capital technologies
shown in Table 3.10 may be noted: (a) the recommended
rates of nitrogen vary from 0 to 150 kg/ha with an average
of 108 kg/ha, weighted according to the area planted to
maize (Table 3.11). The zero value corresponds to maize
planted immediately after turning under alfalfa stubble, in
soils with a high water table. The 150 kg/ha value is for
maize plantings in both irrigated soils and soils with a high
water table, where one or more years have elapsed since the
incorporation of alfalfa stubble; (b) the recommended rates
of phosphorus vary from 0 to 60 kg/ha P20s, with a
weighted average of 32 kg/ha; (c) the recommended plant
densities vary from 30,000 to 60,000 plants/ha, with a
weighted average of 48,530 plants/ha; and (d) the variable
costs of these production formulas vary from 0.02 to 1.46
tons/ha of maize, with a weighted average of ,0.98 ton/ha.
The variable costs of the limited capital technologies
vary from 0.02 to 0.92 ton/ha of maize, with a weighted
average of 0.64 ton/ha. The variable costs of the traditional
and INIA technologies are 0.54 and 0.82 ton/ha of maize,
Three assumptions were made in comparing the limited
capital and unlimited capital technologies of the Project
with the traditional and INIA technologies: (a) the aver-
ages of the experimental yields obtained from 1967
through 1972 are reasonable approximations of the average
yields that will be obtained in the future; (b) experimental
yields, when reduced by 20 percent, are reasonable approxi-
mations to commercial yields; and (c) the areas planted to
maize,,costs of inputs and prices of grain and stover will not
soon change appreciably.
The basic data for comparing technologies were prod-
uced in 125 field experiments, which included fertilization,
plant density, and date of planting variables, conducted on
farmers' fields from 1967 to 1972. These experiments
sampled, with varying degrees of intensity, the 16 produc-
ing systems listed in Table 3.9. The areas plants to maize in
each of these systems were estimated from survey data and
the soils map shown in Figure 1.2. The number of experi-
ments conducted in each producing system, areas planted in
maize, and average commercial yields estimated for the four
technologies are shown in Table 3.11.
Table 3.12 shows the distribution of the 125 experi-
ments among years and among five producing systems. The
three largest systems (1.1.1, 2.1.1, and 3) were sampled
quite adequately in each of the six years. The remaining 13
systems, taken individually, were much less adequately
sampled, either because the number of experiments was

small or the distribution among years was inadequate. As a
group, however, the 13 systems were sampled quite satisfac-
torily. The four production technologies are compared here
for producing systems 1.1.1, 2.1.1, and 3; the combined 13
systems; and the entire Project area.
Net increases in maize yields, A M, were calculated for
the different technologies and producing systems according
to the formula:

AM ='Y-C-T

where Y is the estimated commercial yield, C is the variable
cost associated with a given technology above the cost of
the check expressed in ton/ha maize and T is the yield of
the check treatment (no fertilizer, 30,000 plants/ha). As
shown in Table 3.13, estimated average net increases using
the traditional, INIA, limited capital, and unlimited capital
technologies were 0.74, 1.05, 1.12, and 1.44 tons/ha of
maize, respectively.

TABLE 3.10. The production technologies, together with
their costs, recommended for 16 producing systems in the
Project area since 1972.

technology for:
Producing Limited Unlimited
system capital capital

80- 0-40,000
80- 0-40,000
80- 0-40,000
0- 0-60,000

Cost of the technology** in
tons of maize in the field
Limited Unlimited
capital capital

100- 0-50,000
100- 0-50,000
0- 0-60,000

Averages + 75-15-38,260 108-32-48,280 0.64 0.98

* The three values refer to kilograms per hectare of nitrogen, kilograms
per hectare of P205 (phosphorus) and plants per hectare, respectively.
** This is the total cost of fertilizer (price, transport, application, insur-
ance, interest) expressed in terms of grain, after costs of harvesting,
shelling, sacking, and transport have been discounted. The cost of the
traditional and INIA technologies are equivalent to 0.54 and 0.82 tons
of grain, respectively. If these values and the values in the table are
multiplied by $54.80, the costs of the technologies in U.S. dollars are
obtained. This value for maize in the field is based on a price of U.S..
$72 per ton for maize with 14% moisture, placed at a warehouse of the
National Marketing Agency.
+ Weighted according to the areas planted to maize (See Table 3.11).

TABLE 3.11. The number of experiments conducted in 16 producing systems in the Puebla area, areas planted in maize, and
estimated commercial yields* using various production technologies.

Area planted Project technologies
Producing No. of in maize Traditional INIA Limited Unlimited
system expts. ha Check** technology technology + capital capital

1.1.1 27 10586 0.52 2.05 2.86 2.63 3.80
1.1.2. 16 7072 0.38 1.34 1.88 1.49 2.08
1.2 1 1019 0.68 1.88 2.37 2.37 3.36
2.1.1 18 8874 0.79 2.15 2.82 2.82 3.87
2.1.2 8 3642 0.20 1.09 1.54 1.20 1.97
2.2 4 1852 0.48 1.89 2.54 2.54 3.45
3 '24 22739 1.00 2.56 3.29 3.20 3.64
4 7 2078 1.15 2.04 2.49 2.41 2.71
5.1.1 3 2817 0.88 2.12 2.66 2.58 3.66
5.1.2 5 4355 1.44 2.28 2.65 2.59 3.01
5.1.3 5 5636 0.15 1.26 1.61 1.35 1.62
6.1.1 1 1281 0.34 0.66 1.64 1.60 2.21
6.1.2 1 1963 0.52 1.43 1.88 1.55 1.88
6.1.3 3 2540 0.44 0.90 1.08 1.04 1.04
7.1 1 893 4.00 4.00 4.96 5.41 5.41
7.2 2 2653 1.60 3.62 4.12 4.42 5.14

Total 125 80000
Average" 0.78 2.05 2.67 2.54 3.19

The commercial yield was estimated as 80% of the yield obtained experimentally; expressed as tons per hectare of grain
with 14% moisture.
** No fertilizer; 30,000 plants per hectare.
+ Traditional technology; 50 kg/ha of nitrogen, 25 kg/ha of P205 (phosphorus), 10 kg/ha of K20(potassium) and 25,000
plants per hectare.
INIA technology: 80 kg/ha of nitrogen, 40 kg/ha of P205, and 40,000 plants per hectare.
++ Weighted according to the areas planted in each producing system.

Estimated net increases for one producing system (6.1.3)
were:-0.08, -0.18, 0.05, and 0.05 ton/ha, respectively, for
the traditional, INIA, limited capital, and unlimited capital
technologies. These values were calculated from data
obtained in two experiments conducted in 1969 and one in
1970. The implication of these figures could be that it is
irrational to plant maize in this system. However, due to
the limited amount of data available (and to the probability
that farmers' experiences, covering a much longer period of
time than that studied experimentally, has demonstrated
the profitability of maize production), a tentative recom-
imendation was made of 60 kg/ha N, 20 kg/ha P205), and
30,000 plants/ha.

TABLE 3.12. Distribution of the experiments conducted
in the Project area among years and among five producing

Producing in maize
system ha 1967 1968 1969 1970 1971 1972 Total

1.1.1 10,586 9 7 2 3 4 2 27
2.1.1 8,874 3 3 3 3 4 2 18
3 22,739 2 0 6 8' 4 4 24
13 remaining
systems 37,801 8 3 7 19 15 4 56
all 16
systems 80,000 22 13 18 33 27 12 125

Table 3.14 shows the variable costs, net increases, risks,
and "adequacy indices" for the four technologies used in
systems 1.1.1, 2.1.1, and 3; the combined 13 systems; and
the total Project area. As used here, risk is defined arbi-
trarily in two ways: (a) as the standardized probability
that the net increase in maize yield will be 0.5 ton/ha or
less, and (b) as the standardized probability that the net
increase in maize yield will be 0 ton/ha or less. In the
remainder of this chapter, the first criterion of risk will be
expressed as R (0.5) and the second criterion as R (0) .
One of the definitions of risk as a net increase of 0.5
ton/ha or less of grain was based on 1970 survey data
indicating that an average family consumed 1,546 kg/year
of maize for food, and had an average area of 2.27 ha of
maize. Thus from each hectare in maize, an average of 0.68
ton of grain was needed to feed the family. In this defini-
tion of risk, it was assumed that most of the yield obtained
with the check treatment would be used to cover the fixed
costs of production, and that 0.5 ton/ha to feed the family
would have to come from net increase in yield.

TABLE 3.13. Estimated net increases in tons of maize per
hectare, A M*, using four production technologies in 16
producing systems.

Producing Limited Unlimited
system Traditional INIA capital capital

1.1.1 0.99 1.52 1.51 2.10
1.1.2 0.42 0.65 0.56 0.78
1.2 0.67 0.88 0.88 1.37
2.1.1 0.82 1.19 1.19 1.77
2.1.2 0.35 0.47 0.50 0.67
2.2 0.87 1.24 1.24 1.67
3 1.05 1.45 1.61 1.92
4 0.38 0.59 0.80 0.93
5.1.1 0.73 0.96 0.96 1.53
5.1.2 0.30 0.39 0.39 0.47
5.1.3 0.57 0.64 0.65 0.65
6.1.1 -0.22 0.49 0.50 0.77
6.1.2 0.38 0.55 0.49 0.55
6.1.3 -0.08 -0.18 0.05 0.05
7.1 -0.54 0.14 1.39 1.39
7.2 1.48 1.70 1.91 2.08

Overall** 0.74 1.05 1.12 1.44

The commercial yield obtained with a given technology,
less the check yield, less the variable costs associated
with the use of the technology. The commercial yield is
estimated to be 80%of the experimental yield, ex-
pressed as grain with 14% moisture.
** Averages weighted according to the areas planted in

The value A M 0 is used in the alternative defini-
tion of risk, to examine the case in which the value of the
increase in maize yield is equal to or less than the variable
costs of production. Information is available in Table 3.14
to calculate risk using net increase values other than 0.5 or
0 ton/ha.
In the calculation of risk there is an implicit assumption
that the net increase values belong to a population with a
normal distribution. This hypothesis was tested for tradi-
tional technology used in the entire Project area, and was
not rejected.
For the five systems in Table 3.14, calculations were
made of the mean squares of the net increase values as-
sociated with the following: (a) years, with a degrees of
freedom; (b) total, with n-1 degrees of freedom; and (c)
residuals (sites confounded with the interaction sites x
years), with n-l-a degrees of freedom. In 18 of the 20
cases (four technologies x five systems), the mean square
associated with years was larger than that associated with
sites plus the site x years interaction.. In nine cases, the
difference was not significant; in six it was significant at the
5 percent level; and in three cases it was significant at the 1
percent level. The mean square associated with years was
selected as the estimator of the variance of the net increase
values. This quantity appears in Table 3.14 as s2 AM
The values of risk presented in Table 3.14 were obtained
from a tabulation of areas corresponding to a normal
population with a mean of 0 and a variance of 1. To use
this table, the values 0.5 and 0 were standardized for each
technology and producing system by subtracting the value
of A M and dividing by the appropriate standard deviation,
s AM'
The use of traditional technology in producing system
1.1.1 gave an average net increase in maize production of
0.99 ton/ha, with a risk represented by the number 0.213
for R(0.5) and the number 0.054 for R(0) (see Table
3.14). This level of risk indicates that the net increase will
be 0.5 ton/ha or less in four years out of 20, with one year
having zero or negative net increase. The estimated net
increase using traditional technology in system 1.1.1 was
more than that for system 2.1.1, and similar to that for
system 3. The level of risk in using traditional technology in
systems 2.1.1 and 3 indicates that net increases of 0.5
ton/ha or less can be expected in 6 years out of 20. In two
of the six years, the net increase will be zero or negative for
system 2.1.1, whereas in system 3, three of the six years
will have a zero or negative net increase of yield. These
three systems are the most productive in the Project area
and account for 53 percent of the area planted to maize.
The use of traditional technology in the 13 remaining
systems gave an average net increase of 0.44 ton/ha with a
risk represented by the numbers 0.547 for R(0.5) and
0.195 for R(0). That is, in 11 out of every 20 years a
net increase of 0.5 ton/ha or less can be expected; 4 of
these 11 years will show a zero or negative net increase.
The average net increase for the whole area using tradi-
tional technology was 0.73 ton/ha with a risk represented
by the numbers 0.399 for R(0.5) and 0.206 for R(0).

This means that in 8 out of 20 years the net increase of
yield will be 0.5 ton/ha or less; whereas in 4 of the 8 years
the net increase will be zero or negative.

Average net increases in production per unit of cost,
expressed as kilograms of maize, are shown in Table 3.14
as values of A M/C. These values were 1.83, 1.52 and 1.94
for traditional technolnov in vyetems 1 1 1 2 1 d,,l '.

is defined as R(0), the limited capital technology becomes
slightly riskier in system 1.1.1 than the traditional tech-
nology: 0.065 versus 0.054. With the same definition of
risk, the unlimited capital technology is slightly riskier than
the traditional technology in the total area (aggregated 16
producing systems).

S. J .P..., The unlimited capital technology was superior to the
0.81 for the combined 13 systems; and 1.35 for the total IN technology when compared in terms of net increase in
area. INIA technology when compared in terms of net increase in
yield, or as risk defined as the probability that the net
When compared With the traditional technology, in all increase in yield be equal to 0.5 ton/ha or less, R (0.5), in
five individual and aggregated producing systems, the INIA, the five individual and aggregated producing systems. Using
limited capital, and unlimited capital technologies produced the same criteria, the unlimited capital technology was also
higher net increases in maize yields and lower risks, with better than the limited capital technology in systems 1.1.1,
two exceptions. These two exceptions were in relation to 2.1.1, and 3; whereas in the remaining systems, net in-
risk and not in relation to net increases in yields. When risk creases in yield were larger with the unlimited capital tech-

TABLE 3.14. Variable costs, net increases, risks, and "adequacy indices" for four production technologies used in several
individual or aggregated producing systems.

Individual or cost
aggregated of Net
producing technology increase
system* Technology C AM ** AM s2 + Risk, R + Adequacy indices O of the technologies
(ton/ha) (ton/ha) C AM P( AM 0.5)P( AM 0) la0 Ib Ila lib Ila IIIb

1.1.1 (27) Traditional 0.54 0.99 1.83 0.3804 0.213 0.054 1.00 1.00 1.00 1.00 0.66 0.91
INIA 0.82 1.52 1.85 0.6932 0.110 0.034 2.97 2.45 1.96 1.61 1.26 1.44
Limited Capital 0.60 1.51 2.52 0.9967 0.156 0.065 2.09 1.25 1.88 1.13 1.18 1.37
Unlimited Capital 1.19 2.10 1.76 1.0200 0.056 0.019 8.01 6.06 3.63 2.75 1.91 2.04

2.1.1 (18) Traditional 0.54 0.82 1.52 0.3749 0.304 0.090 1.00 1.00 1.00 1.00 0.41 0.70
INIA 0.82 1.19 1.45 0.3948 0.136 0.029 3.25 4.45 2.14 2.93 0.92 1.13
Limited Capital 0.82 1.19 1.45 0.3948 0.136 0.029 3.25 4.45 2.14 2.93 0.92 1.13
Unlimited Capital 1.30 1.77 1.36 1.1453 0.117 0.049 5.60 3.93 2.33 1.63 1.41 1.62

3 (24) Traditional 0.54 1.05 1.94 1.0190 0.295 0.149 1.00 1.00 1.00 1.00 0.58 0.81
INIA 0.82 1.45 1.77 1.7421 0.236 0.136 1.73 1.52 1.14 1.00 0.92 1.14
Limited Capital 0.60 1.61 2.68 2.0559 0.219 0.131 2.06 1.74 1.85 1.57 1.13 1.32
Unlimited Capital 0.75 1.92 2.56 3.3232 0.218 0.147 2.47 1.86 1.78 1.34 1.34 1.53

13 sys-
tems (56) Traditional 0.54 0.44 0.81 0.2644 0.547 0.195 1.00 1.00 1.00 1.00 -0.10 0.25
(aggegated) INIA 0.82 0.63 0.77 0.1399 0.364 0.046 2.15 6.00 1.42 3.95 0.10 0.56
Limited Capital 0.64 0.68 1.06 0.0918 0.277 0.013 3.05 23.93 2.57 20.10 0.31 0.67
Unlimited Capital 0.99 0.86 0.87 0.5159 0.308 0.115 3.47 3.31 1.89 1.81 0.29 0.65

Overall (125) Traditional 0.54 0.73' 1.35 0.8021 0.399 0.206 1.00 1.00 1.00 1.00 0.22 0.47
INIA 0.82 1.06' 1.29 1.2194 0.306 0.168 1.89 1.78 1.24 1.17 Q.48 0.74
Limited Capital 0.64 1.11' 1.73 1.3319 0.298 0.168 2.03 1.86 1.71 1.56 0.59 0.82
Unlimited Capital 0.98 1.46' 1.49 3.4542 0.303 0.215 2.63 1.92 1.45 1.06 0.72 0.94

* The number of experiments conducted in each system is given in parenthesis.
** The commercial yield obtained with a given technology, less the check yield, less the variable costs associated with the use of the technology.
+ s AM is the estimator of the variance among years, of the net increase values (5 degrees of freedom).
++ The standardized probability of obtaining a net increase equal or smaller than 0.5 and 0 ton/ha.
,_AMi + Ri Ci
SAMt ; II=I Ct; III=AM(1-R)-CR
The index i, indicates INIA, limited capital and unlimited capital technologies; t indicates traditional technology.
0 la, IIa, and liIa refer to the case when R=P( A M 0.5 ton/ha); Ib, lib and IIIb refer to the case when R=P( A M 5 0 ton/ha).
5 Unweighted averages.

nology, but risk was also higher. Using R (0) as the criterion
of risk, the unlimited capital technology was riskier than
both the limited capital and the INIA technologies in all
instances, except in system 1.1.1.
For the five individual and aggregated producing sys-
tems, the unlimited capital technology produced net in-
creases in yield about twice those obtained with the tradi-
tional technology. The risk using unlimited capital tech-
nology was 26 to 76 percent of that using traditional tech-
nology with R (0.5), and was 35 to 104 percent of that
using traditional technology with R (0). The variable costs
of the unlimited capital technology were 1.39 to 2.41 times
greater than those of the traditional technology.
Net increases in yield using the limited capital tech-
nology and the INIA technology were equal for systems
1.1.1 and 2.1.1. The risk using the two technologies was the
same in system 2.1.1, but was higher by 42 percent when
using R (0.5) and by 91 percent when using R (0), for the
limited capital technology in system 1.1.1 (13 percent of
the area). Variable cost of the limited capital technology
was 27 percent lower than that of the INIA technology in
the same system. Compared to the INIA technology in
system 3 (29 percent of the area), the limited capital tech-
nology had a variable cost that was 27 percent less, a net
increase in yield 11 percent higher, and a lower risk factor
that was less by 7 percent using R (0.5), and less by 4
percent using R (0). The same comparison of technologies

for the aggregated 13 systems (47 percent .of the area)
shows a variable cost for the limited capital technology that
was 22 percent less, a net increase in yield that was 8 per-
cent higher and a risk factor that was 24 percent lower
using R(0.5) and 72 percent lower using R(0).
Average net increases in production per unit of cost,
A M/C, were higher using the limited capital, as compared
to the traditional technology in four of the five systems.
This was true in spite of the higher cost of the limited
capital technology. The INIA technology was superior to
the traditional technology in net increase per unit of cost
only in system 1.1.1. The unlimited capital technology was
superior to the traditional technology, using the net in-.
crease per unit of cost as a measure of efficiency, in 76
percent of the area (system 3 and the combined 13 sys-
The "adequacy indices" in Table 3.14 provide additional
criteria for comparing the four technologies. Index Ia
provides a measure of the relative net increase in yield per
unit of risk for R (0.5), and Ib provides a similar measure
for R (0), using traditional technology as a base. According
to Index Ia, the unlimited capital technology in system
1.1.1 is eight times better than the traditional technology.
Indices Ia and Ib, however, do not take into account the
differences in variable costs associated with distinct tech-
nologies. Indices IIa and IIb do incorporate this concept,
and, for system 1.1.1, Index IIa shows that the relative net

TABLE 3.15. Comparison of four technologies, assuming that each was used in the production of 80,000 hectares of maize.

Limited Unlimited
Traditional INIA capital capital

Average yield (ton/ha) 2.05 2.67 2.54 3.19
Total production of grain (tons) 164 211 213 311 203 366 254 844
Average net increase (ton/ha of
grain)* 0.74 1.05 1.12 1.44
Total net increase in grain (tons) 59 204 84 244 89 769 114 821
Total net increase in stover (tons)** 75 457 93 119 86 270 109 245

Value of net increase, A P+ $3,666,928 $5,138,048 $5,402,432 $6,903,960
Fertilizers used:
Ammonium sulphate (tons) 20 000 32 000 29 823 43 003
Simple superphosphate (tons) 9 756 15 609 6 067 12 775
Potassium chloride (tons) 1 333 0 0 0

Total cost of fertilizers, F+ + $2,353,584 $3,550,704 $2,787,552 $4,267,176

A P/F ratio 1.56 1.45 1.94 1.62

The increase is the commercial yield, less the check yield, less the variable costs expressed in ton/ha of grain. The average
net increase is weighted according to the area in each producing system.
** Net increase in stover is the yield with a given technology, less the check yield.
+ Value of the grain in the field was $54.80/ton; value of stover in the field was $5.60/ton. These are market prices less
costs associated with harvesting and marketing.
++ The cost of fertilizer was the market price plus costs of transport, application, interest on loan, and crop insurance.

increase in yield per unit of risk, adjusted for differences in
variable costs, is 3.63 times greater for unlimited capital
technology than for traditional technology. Index IIIa is a
measure of the outcome of a game in which the farmer
plays to win A M and has a probability of 1-R (0.5) of
doing so, but also has a probability, R (0.5), of losing C.
A comparison will next be made of the impact of the
four technologies on production, net increases, and ferti-
lizer consumption, assuming each technology were to be
used on the 80,000 ha of land normally devoted to maize
production in the Project area. The data needed for this
comparison are shown in Table 3.15. Estimated total
production with the four technologies varies from 164,211
to 254,844 tons/year; the value of net increase varies from
$3,666,928 to $6,903,960; and the cost of fertilizers varies
from $2,353,584 to $4,267,176.
Compared with the traditional technology, the INIA
technology would require a 51 percent larger investment in
fertilizers and would produce 30 percent more maize with a
net increase worth 40 percent more. That is, using the INIA
technology, farmers could invest $1,197,120 moreinferti-
lizers and gain an additional $1,471,120. Each additional
dollar spent on fertilizers would yield a profit of $1.23.
Globally, each dollar invested in fertilizers using the tradi-

tional and INIA technologies produces a gain of $1.56 and
$1.45, respectively.
The limited capital technology would require an 18 per-
cent larger investment in fertilizers than would the tradi-
tional technology; would produce 24 percent more maize,
and would yield a net increase worth 47 percent more.
Using the limited capital technology, farmers would invest
$433,968 more in fertilizers than with the traditional
technology, and would gain an additional $1,735,504.
Thus, each additional dollar spent on fertilizers with the
limited capital technology would give a profit or $4.00. In a
global sense, each dollar invested in fertilizers using the
limited capital technology would yield a profit of $1.94
compared to $1.56 in the case of the traditional tech-
The unlimited capital technology compared with the
traditional, would require an 81 percent larger investment
in fertilizers, produce 55 percent more maize, and yield a
profit worth 88 percent more. Farmers would invest
$1,913,592 more in fertilizers with the unlimited capital
technology as compared to the traditional, but could gain
$3,237,032 more. In this case, each additional dollar spent
on fertilizers would yield a profit of $1.69.



Prior to the Puebla Project, it was known that Chalque-
fio and C6nico were the predominant races of maize in the
region. The Mexican Agricultural Research Institute (INIA)
had done some varietal testing, and two hybrids (H-28 and
H-129) were recommended for the area. A limited survey in
the fall of 1966 indicated, however, that most farmers were
growing native varieties. This finding was confirmed by a
farm survey in early 1968 which revealed that, although 15
percent of the farmers had used hybrid maize on at least
one occasion, less than 1 percent of the farmers had grown
hybrids in 1967.
It seemed reasonable to expect varieties that yield more,
particularly in unfavorable years, would be readily accepted
by farmers and would represent an economical way to in-
crease production. Thus, maize improvement research
became an integral part of the Puebla Project. The research
objective was to quickly develop improved varieties that
would yield more than the available hybrids and native
varieties, and that would compare favorably in terms of
grain type, lodging, earliness, and disease resistance.


The maize improvement program consisted of the fol-
lowing activities:

(a) The collection of information from farmers
throughout the Project area to establish farmer
preferences as to grain type, earliness, and other
morphological characteristics.

(b) The collection of outstanding native varieties in the
area. It was expected that some of these might be
useful for immediate distribution, and many would
be valuable as breeding materials.

(c) The testing of promising local varieties and exotic
materials at representative sites throughout the area.
Initially these varietal trials were to identify out-
standing genotypes, both for immediate use and as
breeding materials, and subsequently to compare
the performance of existing and newly produced

(d) The development of cryptic double-cross (S I x Si)
hybrids and S i x double-cross hybrids. The decision
to use this breeding method was based on experi-
ence in other areas indicating that it should be
possible to develop a hybrid by the third year of the
Project that would outyield the parental varieties by
25 to 30 percent. This timetable was dependent on
growing two crops per year, through winter plant-
ings at lower altitudes. Since the proposed life of
the project was only 5 years, it was necessary to
have improved materials available by the end of the
third year, if they were to significantly influence
production within this time period.
(e) The development of open-pollinated varieties
through mass selection. This method was chosen on
the basis of research experience suggesting that in-
crease in yield could be expected, varying from 4 to
10 percent per year. In addition, since farmers
would cooperate in the selection, they would have
improved seed available immediately and could
continue to attain better yielding materials through
their own efforts after the Project ended.

L -i--

Varietal trials were carried out each year at several loca-
tions, to compare the performance of native varieties, im-
proved varieties and hybrids, and experimental materials.


As information was collected from farmers, it became
evident that length of the growing season was a major
concern of farmers in deciding which variety to plant. A
majority of plantings are made in late March, April, and
early May, in soils that conserve sufficient moisture from
the previous rainy season. Farmers use late-maturing varie-
ties for these early plantings. Early-maturing varieties make
up the remainder of the maize planted in late May and June
after the rains begin.
Farmers find a wide range of kernel colors-white, yel-
low, red, blue, and mixes-acceptable for home consump-
tion. For the market, however, whites and yellows are
preferred, since local buyers sometimes discriminate against
reds and blues.
Overall, the maize improvement program has emphasized
the production of high-yielding varieties, that are resistant
to diseases and lodging, for both early and late plantings.

Production of Hybrids

During the summer of 1967, several local varieties were
examined, and the variety Pinto Salvatori was chosen as
germplasm for the production of cryptic double-cross
hybrids. Five hundred crosses of selected plants were made
and the second ears of each of the 1,000 parental plants
were self-pollinated. Because of problems in obtaining suf-
ficient seed of the cross, as well as the self-pollinations, the
program realized only 94 complete sets.
During the winter of 1967-1968, topcrosses were made
at the experiment station of the National Seed Production
Company near Tepalcingo, Morelos, by crossing H-28 and
S lines of several varieties that were outstanding in the
summer varietal trials. Individual plants of the variety were
selfed at the same time they were crossed with 10 to 15
plants of H-28.
The 94 cryptic double-crosses from the variety Pinto
Salvatori and 68 topcrosses from the winter program were
yield-tested in 1968 at four locations in the Puebla area.
Eleven of the cryptic double-crosses and eight of the
topcrosses to H-28 yielded significantly more than did the
best commercial hybrids.
The S I parents of the five best cryptic hybrids were
planted for increase and to obtain more seed of the crosses
at the Tepalcingo station during the winter of 1968-1969.
Also, these 10 parental lines were arbitrarily divided into
two groups to form two composites (A and B). The cross
between these two composites was made in detasseling
blocks and seed was produced for semi-commercial testing
in the Puebla area.
Comp A x Comp B, together with the five best topcros-
ses to H-28 and other promising materials, was tested at
three locations in the Puebla area during the summer of
1969. Selected farmers were given small lots of Comp A x
Comp B for comparison against their local varieties. The
results with Comp A x Comp B were not up to expecta-

During the summer of 1970, Comp A x Comp B was
included in varietal tests at 16 locations and was planted on
a semi-commercial scale at a few sites. From the results
obtained in 1969 and 1970, it was evident that Comp A x
Comp B was not superior to the parental variety, and it was
decided not to promote the use of Comp A x Comp B in
the Puebla area.
The five outstanding cryptic hybrids, the five best
topcrosses to H-28, and Comp A x Comp B, were continued
in the varietal trials in 1971 and 1972. The relative yields,
days to flowering, and lodging percentages of these hybrids
and other promising materials, are given in Table 4.1.
Several conclusions can be drawn from these data: (a)
compared to the parental variety Pinto Salvatori, Comp A x
Comp B yields slightly less, has a slightly longer growing
season, and has the same tendency to lodge; (b) four of the
cryptic hybrids outyield Pinto Salvatori by 5 to 11 percent
(two of these, 113 and 246, lodge less and have about the
same growing season as the parental variety); and (c) the
five topcrosses with H-28 yield from 2 to 7 percent more
than H-28 (topcross 257 yields 7 percent more and flowers
in 3 days less than H-28).

Production of Varieties
Observation nurseries, including 41 composites of earlier
collections from the Puebla area and 18 other promising
materials, were planted at two locations in the summer of
1967. The relative performance of the several entries
provided guidance to select those materials to be used in
the genetic improvement program. Two composites were
formed at Tepalcingo during the following winter: (a) an
early composite by intercrossing Puebla groups 10, 11, 26,
and 30; Chapalote x C6nico; Chalquefio x C6nico; Harinoso
de Ocho x C6nico; Colorado Salvatori; and H-28; and (b) a
late composite by intercrossing Puebla groups 33, 44, and
49; Batin E-CIV; Hidalgo 8 M-CI; Pinto Salvatori; and Blan-
co Salvatori.
During the summer of 1968, mass selection blocks were
planted and carried through the first cycle of selection with
the early and late composites. The first cycle of mass selec-
tion in an opaque-2 composite was also conducted. The
opaque-2 composite was formed by mixing seed that car-
ried the opaque-2 gene from Mexico group 10, Hidalgo 8,
Pinto Salvatori, and Blanco Rubin.

The mass selection blocks were fertilized each year
according to the recommendations of the Project. Plant
densities of 24,000 plants/ha were used in 1968, 1969, and
1970. Densities were changed to 50,000 plants/ha in 1971
as a result of a study at Chapingo indicating that the plant
density in mass selection blocks should be similar to that in
commercial plantings.
Mass selection with the early composite was done at four
locations in 1968, five locations in 1969, and one location
in 1970. Work with this composite was discontinued in
1971 when it became clear that it was not sufficiently early
for late May and June plantings in the Puebla area.

TABLE 4.1. Relative yields, days to flowering (50%of
tassels showing), and lodging percentages of selected maize
varieties and hybrids studied in the area of the Puebla Proj-

( Z d Z < no r,
.? .(D r 0. OW 0

CD n 0


H-28 84 228 100.0 95 4
Colorado Salvatori 26 88 97.5 91 10
Rojo Salvatori 45 96 85.4 87 7
Pinto Salvatori 84 228 103.2 97 13
H-129 82 224 102.6 106 7
H-129 (before Apr. 21) 30 78 106.9 -
H-129 (after Apr. 20) 52 146 100.0 -
H-127 27 92 93.1 101 4
H-125 23 84 102.2 102 4
H-131 19 56 119.5 106 7
CDC 358 14 40 112.7 102 12
CDC 275 14 40 108.7 100 13
CDC 246 14 40 110.6 95 9
CDC 205 14 40 101.7 101 17
CDC 113 14 40 114.1 97 9
Comp Ax Comp B 28 86 102.4 100 13
H-28 x Pue gpo 44-309 15 44 105.0 97 5
H-28 x Colorado 292 15 44 102.2 89 6
H-28 x Colorado 257 15 44 0187.2 92 6
H-28 x Colorado 276 15 44 102.2 91 3
H-28 x Pue gpo 44 333 15 44 105.1 90 3
Comp 1T SMP 12 36 103.2 100 13
Comp 1500 35 92 97.2 100 15
Local variety 24 64 90.3 92 12

* Average yield expressed as a percentage of H-28 (average yield of
H-28 at 84 sites = 5.47 ton/ha of grain with 14% moisture).

Mass selection with the opaque-2 composite was realized
at one location in each of the years from 1968 to 1971.
Mass selection with the late composite was carried out at 22
sites during the years 1968-1972, an average of 4.4 sites/
year. CIMMYT decided in 1972 to discontinue the mass
selection work, as well as other breeding activities. This
decision was based on the assumption that maize breeding
activities could not be conducted successfully on farmers'
The late composite was included in varietal trials in
1970, 1971, and 1972; average yields of the late composite
in 1972 before mass selection and after the fourth cycle
were 6.14 and 6.38 tons/ha, respectively. Apparently, four
cycles of selection produced little or no improvement in the
late composite. Also, as seen in Table 4.1, the late com-
posite, Comp IT SMP, yields the same as Pinto Salvatori, is
slightly later, and has the same tendency to lodge.
In 1972, the opaque-2 composite, after four cycles of
selection, was compared with seven INIA opaques, three

CIMMYT opaques, and three normal hybrids. The opaque-2
composite of the Project produced 4 percent more opaque
grain than the best INIA material and 16 percent more than
the best CIMMYT material. However, it still yielded well
below the hybrids with normal grain (85 percent of H-129
and 70 percent of H-131).

Evaluation of Materials

A total of 163 varietal trials was done in the Project area
during the period 1967-1972. These trials included farmers'
varieties from the Puebla area and similar regions; improved
varieties and hybrids; and experimental materials from
CIMMYT, INIA, the Graduate College at Chapingo, and the
breeding program of the Puebla Project. Separate trials were
conducted for late materials, early materials, and opaques.
These trials were conducted at population and fertilization
levels similar to the unlimited capital recommendations of
the Project.
The relative yields of 21 of the most outstanding mate-
rials are shown in Table 4.1, along with days to flowering
and lodging percentages. Pinto Salvatori is an outstanding
native variety and should be used more widely in the area.
In yielding ability it compares favorably with H-129 and
H-125 and is superior to H-28 and H-127. A recently re-
leased INIA hybrid, H-131, is the highest-yielding material
studied, outyielding Pinto Salvatori and H-129 by about 16
percent. It should be recommended for March and April
plantings in the Project area.
The varietal evaluations summarized in Table 4.1 include
a small sample of local varieties (only the eight collected in
the spring of 1967). A second collection of native varieties
was made in the winter of 1970-1971, including 216 from
Puebla, 20 from Tlaxcala, 9 from Hidalgo, and 4 from
Veracruz. These were divided into early and late materials
and included in evaluation trials in 1971 and 1972.
In Table 4.2, the average yields and days to flowering of
20 of the best late native varieties are compared with Pinto
Salvatori, H-129, and H-131. Pinto Salvatori and H-131
outyielded all the native varieties. On the other hand, the
native varieties outyielded H-129. These findings indicate
that many of the native varieties in Puebla compare favora-
bly in yielding ability to the best improved materials
presently available. (It should be remembered that when a
local variety and a hybrid yield almost equally and are
similar in other respects, the local variety is preferred
because of the expense and other problems associated with
the production and distribution of hybrid seed.)
In Table 4.3, the average yields and days to flowering of
18 of the best early native varieties are compared with the
hybrids H-35E, H-30, and H-28. The materials are arranged
in order of earliness to facilitate the comparison of varieties
with similar growing seasons. Both H-30 and the experimen-
tal hybrid, H-35E, outyielded all native varieties that had a
similar number of days to flowering. H-30 flowered five
days earlier than H-28 and should be useful for May and
early June plantings. H-35E flowered a week before H-30
and might be suitable for mid-June plantings.


The maize improvement program did not meet its goal
of developing higher-yielding materials and putting them
into commerical production by the seventh year of Project
operation. Two of the best cryptic hybrids outyielded the
best materials available in 1967 by about 10 percent.
However, as the parental lines of these crosses yielded
poorly and lodged badly, it was not feasible to produce
these hybrids commercially. The Comp A x Comp B,
formed from the parental lines of the five best cryptic
hybrids, could have been produced at low cost, but un-
fortunately it did not retain the high yielding capacity of
the single crosses.

TABLE 4.;'. Average yields and days to flowering of late
maturing local varieties and introduced hybrids. The values
are averages for seven experiments carried out in 1971 and

Yield of grain Days to
Material with 14% moisture flowering

Pinto Salvatori 5.52 107
Pue. 26 5.45 118
Pue. 66 5.36 107
Pue. 41 5.30 118
Pue. 77 5.30 111
Pue. 108 5.28 108
Pue. 79 5.28 114
Pue. 27 5.21 120
Pue. 67 5.21 106
Pue. 119 5.17 111
Tlax. 145 5.17 113
Pue. 69 5.17 113
Pue. 45 5.12 113
Pue. 4 5.10 105
Pue. 62 5.08 106
Pue. 2 5.07 104
Pue. 59 5.07 108
Pue. 10 5.06 105
Pue. 116 5.06 108
Pue. 141 5.04 112
Pue. 36 5.04 107
H-131 5.60 120
H-129 4.65 121

TABLE 4.3. Average yields and days to flowering of early
maturing local-varieties and introduced hybrids. The values
are averages for four experiments carried out in 1971 and
Yield of grain Days to
Material with 14%moisture flowering

Tlax. 237 2.27 82
Pue. 178 2.35 83
Pue. 153 2.59 84
Pue. 217 2.49 85
H-35E 3.14 86
Pue. 139 2.70 86
Pue. 175 2.52 87
Pue. 214 2.68 87
Pue. 184 2.75 88
Pue. 183 2.80 '88
Pue. 53 2.83 88
Pue. 216 2.90 89
Pue. 159 3.08 89
Pue. 210 2.78 90
Pue. 200 3.26 90
Pue. 86 2.88 91
Pue. 91 2.92 91
Pue. 29 3.07 91
Pue. 195 3.01 92
H-30 3.82 93
H-28 3.60 98

Four years of mass selection in the late composite at a
total of 19 sites produced little or no improvement in
yielding ability. This result is not in accord with the
experiences of many maize breeders and possibly was
influenced by the following considerations: (a) the plant
density in the selection blocks in 1968, 1969, and 1970 was
only about half that used in commercial plantings; there is
some evidence that plants that are outstanding at low
densities are not necessarily superior at high densities; (b)
there were difficulties at many sites in achieving complete
isolation of the selection block, because the adjoining plant-
ings could not be controlled; this may have resulted in the
introduction of undesirable germplasm into the composite;
and (c) the land chosen for some of the selection blocks
was quite variable; this made it difficult to select only those
plants that were genetically superior.
The major contribution of the maize improvement
program has been in determining the usefulness of local and
introduced materials for early and late plantings in the area.
Pinto Salvatori is an outstanding local variety that should
be used more widely for plantings in March, April, and

early May. H-131 yields about 16 percent more than Pinto
Salvatori and is recommended for March and April plant-
ings. H-30 is superior to local varieties for late May and
early June plantings. H-35E shows promise for mid-June
plantings. In general, the maize improvement work demon-
strated that many local varieties are high-yielding when
production conditions are favorable.
Maize improvement experience in the Puebla Project
indicates that the development of improved varieties for a
regional program can perhaps best be achieved in a coop-
erative effort with a nearby research center. The crop
improvement component of the regional program would
have the responsibility of collecting the information that is
necessary to clearly define the characteristics of the im-
proved varieties needed by farmers. The regional program
would assist in the collection of local genetic materials that

might be useful in producing such improved varieties. It
would determine the major conditions used for crop
production in the area and conduct evaluation trials at sites
located so as to adequately sample these conditions.
The crop improvement program at a neighboring re-
search center would have the responsibility for selecting the
materials and methods for producing improved varieties. It
would supervise all breeding activities, both in the regional
program and at the research center. The selfing, crossing,
and selection of materials might be done either at the re-
search center or at appropriate locations in the Project area.
The selection of plants tolerant to moisture stress, for
example, might best be made at appropriate sites in the
Project area. The most experienced personnel available
should participate in any step involving a subjective evalua-
tion of materials.

Field demonstrations were held at harvest time, to show
farmers how yield and net income were increased by using
the new technology. Here the net returns from using the
recommended number of bags of fertilizer are being dis-
cussed with the farmers.




Technical assistance to Puebla Project farmers began in
early 1968, when new maize recommendations were devel-
oped based on the 1967 research results and complementa-
ry data. This new maize technology brought several changes
for its users, including (a) higher investment in fertilizers-
the new fertilizer recommendation in 1968 cost about 130
percent more than the average fertilizer treatment used in
1967 by the 70 percent of the farmers-who fertilized their
maize; (b) purchase of individual fertilizer materials instead
of a formula, and the mixing of the materials at home; (c)
application of a part of the fertilizers at planting time and
the remainder at the second cultivation, instead of applying
all the fertilizer at the first cultivation; (d) use of higher
plant densities-50,000 plants/ha-instead of the 15,000 to
25,000 used earlier; and (e) control of weeds, with more
complete and timely methods, plus control of the rose
chafer at flowering time when necessary.
Although these changes were largely quantitative, they
did imply wide-reaching changes in farm management and
farming practices-for the Pueblan farm families.
The central aim of the technical assistance program was
to provide every possible assistance necessary to enable the
farmers to use the new technology effectively. Technical
assistance agents sought rapid adoption of Project recom-
mendations by concentrating on: (a) providing the farm-
ers with information about the Project, including how the
new recommendations were developed and the several
components of the new recommendations; (b) assisting the
farmers in obtaining credit and in arranging for fertilizers;
(c) instructing the farmers in the most efficient ways to use
the recommendations; and (d) collecting information from
the farmers about obstacles limiting their use of agricultural
services in the area, transmitting the information to the
members of the Project team and to representatives of
service agencies, and assisting in finding ways to overcome
the difficulties.


Location of the High-yield Plots

A demonstration program was planned for 1968 so that
the farmers could have a first-hand look at the advantages
of the new maize technology. Plans were made to locate

"high-yield" plots throughout the western three-quarters of
the Project area. Initially, the plans called for the custom-
ary procedure used for locating demonstrations; that is,
choosing of highly accessible locations with good soils,
where the largest possible number of farmers could see the
plots. This approach implied that the fields should be
located first, and the owners then convinced to participate.
However, the experience obtained by the evaluation team
in early 1968 indicated a change in strategy. The team
encountered negative attitudes and, in some cases, hostility
among many farmers; thus plans were revamped to work
through the existing power structure in each community.
The political administrative unit in the Puebla area is the
municipio or county, each of which has a principal village
and several ancillary population units or communities. The
municipal president and other municipal authorities live in
the principal village with auxiliary authorities, responsible
to the municipal president, residing in each of the com-
As a first measure, the Project staff began to contact the
municipal presidents and explain the Project and its goals.
These initial visits provided a brief description of the
Project, using the report prepared for the first annual meet-
ing, a map showing the locations of the experiments
conducted in 1967, and a list of the cooperating farmers.
This basic information was attached to an official letter of
presentation signed by the General Agent of the Ministry of
Agriculture, the State Director of Agriculture, and the
coordinator of the Project. The letter explained the re-
sponsibilities of the municipal authorities and the impor-
tant role they would play in developing the Project.
The presidents were asked to arrange general meetings of
all the municipal authorities, so that full information could
be provided about the Project and the work plans for 1968.
Such meetings were held in all but one of the municipios in
the western three-fourths of the Project area.
During the first meeting with the municipal authorities,
careful explanation was made of what the Project could
provide and how the farmers might cooperate. At the
completion of each meeting, the participants were asked
which farmers in the locality might be interested in the
Project. The authorities usually asked for time to return to
their villages to explain the Project and find out who might
be interested. In a few cases, the local authority himself was
ready to participate and to initiate Project work in his

The next step was to schedule a series of meeting with
farmers in the villages where authorities had expressed some
interest. The local authorities took the initiative in organ-
izing the meetings and inviting the farmers. At these meet-
ings, Project technicians explained the Project and sug-
gested how the farmers might participate by using the new
recommendations in a part of their maize plantings. A total
of 31 such meetings were held.
The farmers learned that they would have to provide the
fertilizers and labor, and the Project technicians would
assist in the field operations to assure that the recommenda-
tions were used correctly. For those who did not have
money to purchase the fertilizers, help was offered in ob-
taining credit from a private or official agency. After all
aspects involved in using Project recommendations had
been explained in detail and discussed at great length, a
small group of farmers gradually took the initiative. These
farmers generally had two characteristics: they were (a)
responsible workers of their land with a desire to progress,
and (b) persons whose moral character and influence were
amply recognized in the community.
In some communities, many farmers wished to partici-
pate; in such cases, the final selection of participants was
made by visiting the possible sites. Generally, no less than
two and no more than five sites were selected in each com-
munity, but in a few cases there were more than five. There
were 25 sites around one village, and eight sites at another,
due principally to the enthusiasm of the farmers and the
fact that the village land was very extensive. There were
only two instances where farmers were accepted and later
withdrew; these withdrawals were due to objections of the
wives, principally because the husband was planning to
obtain fertilizer on credit and they objected to going in


A total of 141 high-yield plots, varying in size from 0.25
to 1.0 ha, were established by 103 farmers. Each farmer
was given information about the availability of credit, the
interest rate, and what the role of crop insurance could be
in reducing risks from natural causes.
In 1968, 60 percent of the farmers who participated
were financed by Agronomos Unidos, a private fertilizer
distributor. An additional 20 percent of the credit was
provided by the Agricultural Bank of the South, and 20
percent of the plantings were self-financed by the farmers.
Credit was provided at an interest rate of 1-1/2 percent/
month. The credit was extended for 9 months, sufficient
time to cover the long growing season and allow the farmer
to harvest and sell enough maize to repay the loan.
The letters of credit which the farmers signed on receiv-
ing the fertilizers were prepared in two ways: most of them
showed only the amount of the loan and the rate of inter-
est; a few indicated the total amount of the loan plus
interest. In those cases where the interest was calculated in
the original loan agreement, and the farmer paid before the
9 months were up, he received a cash refund for interest
corresponding to the difference between 9 months and the

actual loan time. This turned out to be an agreeable surprise
with good will resulting for the distributor who provided
the credit.
In contrast, there was occasional friction when the farm-
er arrived to pay his loan with the understanding that only
the principal was to be paid as indicated in the letter of
credit. In such cases, when the interest was calculated, the
farmer often did not have enough money on hand to make
payment. One such farmer considered the interest a fraud.
Bad feelings often occurred, even among those farmers who
understood that the credit terms were very favorable
compared to local lenders and had simply forgotten to
calculate the interest. Because of these experiences, it was
decided that the total amount of the loan, principal plus
interest, should be stated in the letter of credit, whenever

Crop Insurance

After the plantings had been made, the crop insurance
agency insured them. This was an experimental operation
for the insurance agency, because their usual procedure was
to insure plantings of only 5 ha or more. The plots financed
by Agronomos Unidos varied from 0.25 to 1.0 ha, and at
the outset it was difficult for the insurance agency to
include them. However, the risk aspects of rainfed plantings
were of special interest to the Project, and the participation
of the crop insurance agency was ultimately arranged. The
insurance agency made the necessary inspections of the
plantings, and discarded 14 plots that were considered
unacceptable. The remainder were fully insured. According
to the yield levels obtained at the end of 1968, the insur-
ance agency had no indemnifiable losses whatsoever due to
hail, drought, wind, frost, and other risks covered by the
program. There were reductions in yield due to these
causes, but none that would require payment under the
insurance regulations. The maximum coverage was for a
value equivalent to 1.1 tons/ha grain.

Planting and Care of the High-yield Plots

The high-yield plots were planted on dates decided upon
by the farmer cooperators. The plantings were used as
demonstrations, and the neighboring farmers were invited
to watch the procedures. In some cases, the cooperating
farmer made the fertilizer mixture several days before
planting; in others, the mixing was part of the demonstra-
The farmers were shown how and when to apply the
mixture so that the fertilizer would be evenly distributed at
the bottom of the furrow. A convenient local measure was
found for calibrating the fertilizer distribution a 1-liter oil
can. When this can was filled to about one finger below the
top and distributed over 20 meters, the appropriate amount
of the mixture was applied. For rapid measuring, a
20-meter wire was used to locate stakes at 20-meter inter-
vals along the row.

The program to promote
farmer use of the new maize
technology was initiated in
1968, with 103 farmers par-
ticipating with 141 high-yield
plots. Eighty percent of these
farmers were provided credit
by a fertilizer distributor and
an official bank. Here farmers
are seen signing loan agree-

The traditional planting rate for maize required a full
step distance between hills; however, the new planting rate
was demonstrated in terms of a distance between hills of
about one-half step. The higher population required learn-
ing a new rhythm of planting-inserting the shovel, opening
and covering twice as many holes per hectare.
To assure an optimum population density, the farmers,
at first, were taught to overplant and then thin to the
desired 50,000 plants/ha. In this way, the population could
be assured in spite of soil insects, inadequate germination,
and other factors. However, the fertilizer applied at plant-
ing time caused vigorous early growth. When told that it
was time to thin the plants, the farmers frequently replied:
"Here I have one of the most beautiful plantings of maize
that I have ever grown, and you want me to pull out some
of the plants." To them, pulling of the superfluous plants
was a destructive act. As a result, before the planting season
was far advanced, it was decided to reduce the planting rate
and eliminate the thinning operation.
The technicians kept in contact with the high-yield plots
throughout the growing season. As the plantings were
completed, attention was given to weed control, and where
necessary, to control of the rose chafer. Demonstrations for
neighboring farmers were held at the second cultivation
when the second fertilizer application was made. The
visiting farmers learned which fertilizer to apply, and how
much, as well as how to keep fertilizer out of the bud to
avoid damaging plants.

Result Demonstrations

Local demonstrations were held just prior to harvest at
15 of the high-yield plots, and neighbors and farmers from
adjoining communities were invited via local sound equip-

ment, printed circulars, and posters. Attendance ranged
from 11 to 75 farmers.
The demonstration consisted of three parts: (a) the
technical assistance agent's explanation of the Puebla
Project, (b) the cooperating farmer's report of the practices
used in the high-yield plot, and (c) open discussion led by
the farmer and technician. An interesting aspect of the
discussion was the obviously greater self-confidence felt by
the visiting farmers when raising questions and making com-
ments to the farmer-demonstrator.
Two regional demonstrations also were held just before
harvest at strategic locations where both a high-yield plot
and an experiment could be seen. Farmers with high-yield
plots in each locality were asked to organize the event. In
meetings with these farmers to plan the demonstration, two
aspects were noted: (a) the farmers lacked confidence in
their ability to plan and carry out a demonstration, and (b)
they thought that no one would attend. They felt that the
technical assistance agents should make the decisions. The
technicians, however, encouraged the farmers and insisted
that the farmers handle the arrangements.
The organizational approach which evolved was to name
a committee of the most enthusiastic farmers with the
formal title: Committee for Organizing the Agricultural
Field Day. The committee took charge of: (a) inviting the
authorities, both of the federal and state governments, (b)
inviting the neighbors, (c) naming a person to receive each
of the groups as they arrived from the different communi-
ties, and (d) naming members to look after the smooth
functioning of the demonstration to assure that there
would be an atmosphere of hospitality.
The technical assistance agents invited farmers from
other parts of the Project area, using personal contact, a
poster, a printed circular, plus personal invitations to all farm-

ers with high-yield plots. General attendance was good at
both events.
Through organization of the events, the farmers gained
experiences of lasting value, plus confidence in their ability
to conduct demonstrations of this type. Attendance was
greater than they had imagined possible, and often included
farmers from the more distant villages.
Other demonstrations were held throughout the growing
season for representatives of various Mexican state and
national institutions, including the Secretary of Agriculture,
the Governor of Puebla, state directors of agriculture,
directors of the official banks, and many other profes-
sionals interested in the Project. There were also numerous
visitors from Latin America, Europe, and the United States.

Printed Matter and Audio-visual Aids

In meetings with farmers it was impressive to see that
even those who were barely literate took notes on the
recommendations on scraps of paper. Mimeographed and
printed materials were prepared to ensure accurate record-
ing of the information.
At the end of 1968, a pamphlet was published with
Project recommendations for increasing maize yields enti-
tled: "Would You Like to Increase Your Maize Yields? "
The text was minimal and essential data were shown in
illustrations. Thus, farmers attending a meeting could first
hear the recommendations and then take home a folder
containing the same information.
During the 1968 maize growing season, farmers in the
region also played a central role in the filming of a 16-mm
color movie for use in promotional activities in subsequent
years, entitled: "Would You Like to Increase Your Maize
Harvest? "

:j iJ


Pamphlets were prepared with the information farmers
needed to use project recommendations correctly. The text
was kept to a minimum, and the essential data were pres-
ented in illustrations.
A 16 mm. film in color entitled "Would You Like to
Increase Your Maize Harvest?" was produced in 1968. The film
has been very useful in demonstrating to farmers exactly
how to obtain higher yields of maize. It also serves as an
attraction to bring farmers together to discuss common
problems of credit and input availability.

-" .

.J' ri~i24


The Project began its 1969 program on an optimistic
note: successful contact had been made with the social
network of the farming community in 1968, additional
research results were available, finances had been obtained
to expand the technical team of the Project, and the banks
and fertilizer distributors were ready to expand credit to
make fertilizer more readily available Thus, the Project
team and service institutions decided to attempt to extend
the use of the improved technology to a total of 10,000 ha
operated by about 5,000 farmers.
As in 1968, promotional activities were concentrated in
the western three-fourths of the Project area. This region
was divided into four zones (Fig. 1.2) and a technical as-
sistance agent was given responsibility for each zone.

Organizing Groups

If the four technical assistance agents were to provide
guidance to 5,000 farmers in using the new maize recom-
mendations, it was clearly necessary that the farmers be
organized into groups. Beginning in early 1969, the tech-
nical assistance agents began to hold meetings in the villages
to promote the formation of groups. These meetings were
often organized through farmers who had participated with
high-yield plots in 1968. Many of these farmers were al-
ready aware of the advantages of working together in
groups, particularly because of past problems encountered
in arranging for the transportation of small quantities of
fertilizer at a reasonable price.

A typical meeting began with an explanation of the
Project and presentation of the results obtained by farmers
with high-yield plots in 1968, followed by a showing of the
locally produced color film: "Would You Like to Increase
Your Maize Harvest? ". The film was extremely useful in
demonstrating how higher yields could be obtained. It lent
credibility to the recommendations by its careful documen-
tation of 1968 successes, using local names and places.
About halfway through the film, projection was stopped
and questions encouraged. During this intermission, a
mimeographed map of the region was distributed, that gave
recommendations for each community. Thus, each farmer
could identify his own land, and, on the back of the sheet, find
the specific fertilizer recommendation. The movie was
shown 71 times in 59 villages, with a total attendance of
4,570 in 1969.
After the advantages of the new maize technology had
been amply discussed at a meeting, and farmers had expres-
sed an interest in using it, the technical assistance agent
suggested that the farmers consider forming a group. He
stressed that organization would permit: (a) ready access
to information about the recommendations and assistance
in using them correctly, and (b) easier arrangements for
credit and fertilizers.
Two or more meetings and lengthy discussions were
sometimes necessary before the farmers were convinced of
the advantages of organized action and decided to form a
group. Once the group was formed, the members elected a
representative and proceeded to discuss operational
procedures, acceptance of new members, credit arrange-
ments, etc. The technical assistance agents devoted most of
their time to the formation of the groups, and to assisting
them in increasing their membership, learning about the
Project recommendations, and arranging for credit and

During the cropping season interchanges among farmer groups were organized. A representative of the host group welcomed
the visitors and explained how the farmers in his group were trying to improve their crop production. Then the hosts and
visitors made a walking tour of high-yield plots in the immediate vicinity.

Radio Usage

A radio program about the Project was initiated in
March 1969 over a local radio station that transmits to all
the Project area. The program was aired Sunday mornings
from 7:30 8:00-one of the most convenient hours for
farmers, according to data collected by the evaluation team.
Individuals and groups were advised of the program via a
printed flyer, which included the topics, the radio station,
the hour, and the date. Farmers were encouraged to tell
their neighbors of the program.
The radio program content included recommendations
and news notes about happenings of the moment. For exam-
ple, farmers were notified that fertilizers had arrived at
some location, that farmers who had their land prepared
should mix their fertilizers to be ready for planting, that
they should mix simple super phosphate and ammonium
sulphate in certain proportions and take certain precautions
to preserve it. Basically, the program attempted to provide
specific technical information about matters of current
interest to the farmers. In addition, popular local music was

Supervision of the High-yield Plots

In all villages with one or more organized groups, a
demonstration was held at planting time to instruct the
farmers in the new fertilization and planting techniques. As
in 1968, the technical assistant agent first showed the farm-
ers how to distribute the fertilizer and place the seed. Then
all farmers attending were invited to participate in the
During the 60 days or so following the emergence of the
maize, the technical assistance agents accompanied the
members of the organized groups on field inspections of
their high-yield plots. The technicians called attention to
any deficiencies in the way the fertilizer had been applied,
distance between hills of maize, weed control, etc., ex-
plaining how such factors could reduce yields. It was
emphasized that greater care in employing the new tech-
nology would enable the farmers to realize higher yields
and net income.

Interchanges Among Farmer Groups

During the summer of 1969, after most of the maize had
received the last cultivation, the technical assistance agents
began to talk with several outstanding groups of farmers
about organizing a tour of their high-yield plots and inviting
farmers from other parts of the area to attend. In general,
the idea was received with enthusiasm, and 11 of the groups
proceeded to organize such events.
The members of the group hosting the tour decided
when it should be held, how it should be organized, and
which groups should be invited to attend. The technical
assistance agent provided information about possible groups
to invite, and assisted in delivering invitations to the groups.

The local farmers received the visitors at a convenient
location for beginning the tour. The representative of the
host group welcomed the visitors and explained what the
farmers in his group were doing to increase their maize
production. Hosts and visitors then made a walking tour of
several high-yield plots in the immediate vicinity. The
owner of each plot gave a short explanation of the practices
used in his planting. The other farmers were encouraged to
ask questions, offer suggestions, and tell about their own
maize production practices. As opportunities arose, the
technical assistance agent would point out examples of the
correct or deficient use of the several components of the
improved maize technology. Refreshments were usually
served by the hosts at the close of the tour.
These events, referred to locally as interchanges among
farmer groups, provided opportunities for farmers from
different parts of the Project area to exchange experiences
and ideas about a wide range of subjects. While the tour was
in progress, the conversation was usually centered on ques-
tions related to the production of maize, beans, and other
crops. During meals and afterward, the farmers frequently
broke up into small groups and discussed a variety of farm-
ing and non-farming activities. A total of 570 farmers from
35 villages participated in these interchanges in 1969.

Result Demonstrations

The success of the demonstrations held just prior to
harvest in 1968 prompted planning in early 1969 to hold
six regional demonstrations in the fall of 1969. Convenient
locations were selected for the demonstrations, and several
agronomic trials plus a high-yield plot were installed at each
As in 1968, the local farmers organized the field days.
Project technicians explained the experiments and the
importance of the findings. The owner of the high-yield
plot described the practices that he used and gave a benefit:
cost analysis of the operation. Approximately 1,200 farmers
attended the six regional demonstrations.

Beginning in 1969, technical assistance agents have encour-
aged farmers to organize into groups. This has facilitated
the flow of information on technology to the farmers and
has enabled them to arrange for credit and fertilizers more

*r- :-



The technical assistance program was expanded in 1970
to encompass the entire Project area. A fifth technical
assistance agent was assigned the responsibility for Zone V,
the eastern part of the area.
The first 3 months of 1970 were devoted to an intensive
campaign to increase the number of farmers using Project
recommendations. Meetings were held in the villages
throughout the area to explain Project recommendations
and show the results obtained in previous years by farmers
using the new technology. Again the Project-produced film,
"Would You, Like to Increase Your Maize Harvest? ", was
very useful in this promotional effort and was projected in
116 communities with an attendance of 9,9.00 farmers.
The farmers were encouraged to form new groups and to
increase the membership of groups already functioning.
Groups with very large membership were urged to divide
into smaller groups, to simplify administration and allow
more farmers an opportunity to participate in a leadership
capacity. Operating procedures of the different credit
agencies were explained to the groups and they were as-
sisted in presenting their requests for credit and in comply-
ing with the requisites of the agencies.
During the maize planting and vegetative development
periods, the technical assistance agents concentrated on
helping the farmers to use the new technology correctly. As
in previous years, demonstrations were held at planting
time and field inspections of high-yield plots were made
during the early part of the growing season. During the
summer, there were eight interchanges among groups of
farmers, with 610 farmers from 62 communities participat-
Farmer meetings continued to be held throughout the
year to provide information on crop production practices,
the organization of groups, the operating procedures of
service agencies, etc. Approximately 500 farmers were as-
sisted in attending a field day at the Chapingo research
center of the National Agricultural Research Institute.
Regional demonstrations were held in Zones II, III, IV, and
V just prior to the maize harvest, with an attendance of
1,300 farmers.
By early 1970, the technical assistance agents were
beginning to receive requests from the farmers for informa-
tion and guidance in improving their production practices
for crops other than maize, particularly from farmers who
had used Project recommendations the previous season.
One of the first requests was for recommendations for the
production of beans. This request was transmitted to the
rest of the Project team and research on bean production
was initiated by Project technicians during the 1970 season.

In 1971, the technical assistance program continued to
promote greater use of Project recommendations by the
farmers, to instruct the farmers in the correct use of the
new technology, to assist organized groups in arranging for
credit and fertilizers, and to provide information on a vari-
ety of subjects of interest to the farmers. Project recom-
mendations were available in 1971 not only for maize, but
also for bush beans and alternative crops for maize (in years
when plantings are delayed until late June or early July).
A total of 192 meetings were held in 91 villages with
3,686 farmers in attendance in 1971. Movies were projected
for 1,576 farmers in 60 villages. Ninety-three demonstra-
tions of planting techniques were held in 75 communities,
with 1,389 farmers participating. Interchanges among
groups of farmers were arranged in two communities. Radio
programs were broadcast each Sunday morning at 7:30
from March through December.
During 1971, the farmers continued to bring pressure on
the technical assistance agents to assist them with a variety
of production activities. These requests were communicated
to the other members of the Project team. Most of the
requests, however, required resources and expertise not
available in the Puebla Project, which was funded primarily
to increase maize production. Thus, the technical assistance
agents began to search for ways to mobilize resources from
outside the Puebla Project for use in responding to the
farmers' requests.
As a result of this work, short courses on the manage-
ment of fruit orchards were organized and were attended
by about 1,000 farmers in six communities. These courses
were presented by specialists in fruit culture, employed by
the Mexican state and federal governments. Specialists of
the National Extension Service also participated in demon-
strations held at four locations to show how small trench
silos are used to preserve maize stover in the form of silage.


As in past years, the technical assistance program fo-
cused primary attention on increasing the efficient use of
Project recommendations. A total of 382 meetings were
held in 107 villages, attended by 7,875 farmers. There were
187 demonstrations of the planting techniques and the
second application of nitrogen in 86 communities, with
3,121 farmers participating. Eight regional demonstrations
were held just prior to the maize harvest, with a total
attendance of 771.
The technical assistance agents continued to assist farm-
ers with other activities whenever possible. Farmers in 58
villages were assisted in constructing 119 trench silos for
the preservation of about 1,280 tons of stover. Farmers in
Zones I and II were given help in arranging for 3,600 fruit
trees and establishing 12 ha of orchards. Women in a few
villages were aided in acquiring sewing machines and organ-
izing sewing centers.


The technical assistance program in 1973 continued to
center attention on increasing the adoption of Project
recommendations. These recommendations, however, now
included packages of production practices for the maize-
pole bean association as a result of research conducted
during 1970-1972. The results obtained in many exper-
iments had shown conclusively that net income from the
association could be significantly greater than from either
maize or beans grown alone. Thus, the technical assistance
agents began to tell the farmers about the advantages of the
new technology for the maize-bean association, encouraging
them to try the association on a part of their land.
Several obstacles were encountered, however, that limi-
ted farmer use of the new recommendations for the maize-
bean association. The information that previously had been
available to the service institutions recommended that
maize and beans should be grown alone, rather than in
association. Thus, the crop insurance agency was not pre-
pared to insure the association, and the official credit banks
could not authorize credit for farmers who wished to grow
it. The Project team arranged discussions with the repre-
sentatives of the banks and crop insurance agency, and
explained the research results that clearly demonstrated the
advantages of the association. The representatives of the
institutions were convinced by the research findings, and
modified their operating procedures so that credit and
insurance were available for the maize-bean association.

A further difficulty was presented by the farmers them-
selves. They objected to the recommended plant density for
beans-60,000 plants/ha-contending that it was too high
and would result in severe lodging before maturity. The
Project agronomists agreed that their research data on plant
density for beans was not conclusive and that it should be
investigated further. They insisted, however, that farmers
use the recommended seeding rate for beans in at least a
few rows of their maize-bean associations.
Although these problems greatly reduced the effec-
tiveness of the campaign promoting the new technology for
the maize-bean association, it was possible to get farmers to
use the new recommendations on small plots at many sites
throughout the area.

Use of Agua Ammonia

A new source of fertilizer and credit became available in
1973 to farmers in the Puebla area who made use of Project
recommendations for maize. Guanomex, the decentralized
federal agency responsible for the production and distribu-
tion of chemical fertilizers, was interested in finding a way
to make fertilizers available to small farmers at a lower cost.
Guanomex decided that this objective could be achieved,
using agua ammonia as the source of nitrogen. They offered
to provide agua ammonia to farmers, on credit, along with
the applicators, other equipment, and technical assistance
required for its use. Ammonium phosphate, 18-46-0, also
was made available for application at planting time to
farmers who planned to apply phosphorus.

In 1973, Guanomex pro-
moted the use of aqua ammo-
nia by small farmers in the
Puebla area. Although there
are problems in the design of
the applicator yet to be re-
solved, there is interest in this
source or nitrogen, because of
its lower cost.

The Puebla staff studied the proposal of Guanomex and
decided that the potential advantages for the small farmers
in Puebla of having this additional source of nitrogen avail-
able at a lower cost outweighed the risks involved in moving
ahead with a technology that had not been tested locally.
Therefore, the technical assistance agents began meetings
in early April to inform farmers of the availability of the
new source of fertilizer and credit, and to explain that the
cost of nitrogen in the form of agua ammonia was expected
to be about 60 per cent of that of solid materials. They also
described the characteristics of agua ammonia and the
precautions to observe in its application, etc. Although
many farmers were not convinced that agua ammonia was
equal to the solid sources of nitrogen, they were attracted
by the lower cost and seemed confident that it must be
satisfactory, since the technical assistance agent was recom-
mending it. More than 2,000 farmers signed up to use agua
ammonia on some 5,000 ha of maize.
Guanomex made a horse-drawn applicator available in
late April for testing in the Puebla area. It turned out to be
almost impossible to handle the applicator in the field,
because of its excessive weight and high center of gravity.
The farmers were invited to offer suggestions on how to
improve it Several farmers agreed to assist in redesigning the
applicator. Within a short time, a much lighter, better
balanced, applicator was developed, although still not
totally satisfactory.
It was late May before the redesigned applicators were
available to the farmers, and the plant for the production of
agua ammonia was in operation. By that time, most of the
farmers who had signed up to use agua ammonia had found
it necessary to arrange for solid materials in order to make
the sidedressing application of nitrogen at the proper time.
Guanomex had provided many of these farmers with urea
and ammonium sulphate through two of its local distribu-
tors. In total, agua ammonia was used by about 250 farmers
on approximately 500 ha.

More Effective Group Action

When Project technicians began to promote the organiza-
tion of farmer groups in 1969, they expected that these
groups would gradually develop into strong farmer organ-
izations, with the capability of taking the leadership in
finding solutions to many of their problems. By early 1972,
such development had not occurred. Most farmers looked
upon the groups solely as an instrument for obtaining credit
and fertilizers. Once this was accomplished, they had little
interest in meetings or other group activities until it was
again time to arrange for credit.

As Project technicians and advisors explored ways of
developing the effectiveness of the groups, they were
acutely aware of their lack of experience in such work.
Thus, arrangements were made for a sociologist with many
years of experience in organizing small farmers in Mexico to
devote a part of his time to providing technical assistance to
Project staff.

A new strategy for working with the farmer organiza-
tions was not adopted until mid-1973. Each of the tech-
nical assistance agents then began to hold general meetings,
inviting the members of the several groups in his zone. The
technician presented the proposition that many problems
prevented the farmers from improving their agricultural
production, net income, and general welfare. He suggested
that the farmers themselves could best resolve these prob-
lems. He pointed out that people like himself and represent-
atives of the service agencies could help, but, in order for
their help to be effective, the farmers would have to par-
ticipate more actively in deciding what needed to be done
and how to do it.

Several general meetings were held over a period of
several weeks at which the farmers discussed the problems
which they felt to be the greatest obstacles to progress.
Gradually, they were able to define a small list of problems
that were most pressing, and, of these, the one which they
felt should receive top priority.

The next step was to form a new organization of those
farmers interested in working together to resolve common
problems, and, specifically, the problem they had given top
priority. These new organizations are still in the process of
development and consolidation.

The Union of Progressive Maize and Bean Farmers of
Zone III is an example of these new farmer organizations.
This organization held its first meeting in August 1973 and
has been meeting regularly every two weeks. It decided to
give top priority to finding means for members to obtain
chicken manure at a lower cost. After a few weeks of study,
the Union discovered that by eliminating the mark-ups by
two middle men-the truck driver and the administrator of
the chicken farm-the current price for manure (about
$112/truck load) could be cut in half. The Union began
negotiating directly with the owners of the chicken farms
to obtain this better price.
Although the Union has made some progress in its ef-
forts to make manure available to its members at a lower
cost, it has encountered many obstacles. Its lack of legal
status, for example, has prevented the Union from negotiat-
ing long-term contracts with the manure producers. It is
expected that this problem can be resolved through legaliza-
tion as a Civil Society.
In 1973, the Union had 95 members from nine villages,
with a Governing Board composed of 18 farmers, who are
the old group coordinators and their alternates from the
nine villages. The affairs of the Union are administered by a
president, secretary, and treasurer and their alternates.

Farmers in organized groups periodically hold meetings
with the technical assistance agent. These meetings provide
opportunities for the farmers to obtain information about
production technology, credit, crop insurance, and other
matters of interest to them.





The new maize recommendation promoted in 196&was
used by 103 farmers located throughout most of the Pro-
ject area. Their average yield was 3.98 ton/ha on a total of
76 ha. These results were extremely favorable, and it was
decided that the new technology should be promoted as
rapidly as possible among all farmers in the area.
In 1968, the technical assistance agents had worked
directly with individual farmers. With an estimated 43,300
farm operators in the area and only four technical assist-
ance agents and their assistants available in 1969, it was
clearly necessary to find a way to work with groups of
farmers, rather than individuals. In addition to facilitating
the flow of information to and from farmers, organized
groups of farmers could better: (a) arrange for credit,
inputs, and other agricultural services; (b) encourage favor-
able change in the operating procedures of the service
institutions; (c) secure more favorable prices in the purchas-
ing of inputs and the marketing of produce; and (d) plan
and conduct projects for community improvement.
A review of organizational experiences in other parts of
Mexico and other countries with many kinds of farmer
organizations, produced no particular organizational model
that seemed appropriate for the farmers in Puebla. As a
beginning, therefore, it seemed advisable to: (a) acquire as
much information as possible about farmers' experiences
with, and attitudes toward, group action; (b) provide farm-
ers with information about the advantages of working
together; and (c) assist farmers in organizing in the ways
they felt most appropriate.


The politicalunit in the State of Puebla is the municipio,
or county. It is governed by a council consisting of a pres-
ident, secretary, and treasurer. The council is usually
housed in a central building in the principal town in the
municipio. Most of the legal and administrative actions
affecting the citizens of the various communities in the
municipio occur in that building, including: payment of
property taxes, civil wedding ceremonies, regrestration of
births and deaths, etc. Each village in the municipio has an
auxiliary council with the responsibility for less important
transactions. This type of organization was introduced by
the Spaniards during the Colonia Era and has undergone
few changes.

Many of the villages in Puebla are ejidos. These are com-
munities of farmers who received land from the government
as a result of the agrarian reform. (The title to the land
remains with the ejido. The ejidatario retains use rights to
land within the ejido without paying rent as long as he
farms it, and at the time of his death may will these rights
to a member of his/her immediate family.) The highest
authority within the ejido is the comisariado, or executive
committee. Its principal function is to represent the interest
of the ejidatarios in their relationships with higher authori-
ties and service institutions. A second function is to partic-
ipate in political actions at the community level and,
through the Agrarian Community League, at the regional
and national levels. The comisariado consists of a president,
secretary, treasurer, and vigilance committee.
Credit societies were functioning in several ejidos at the
time of the 1967 survey. An ejidal credit society consists of
those ejidatarios who wish to receive credit from the
National Ejidal Credit Bank, an official credit agency
formed with the express purpose of providing credit to the
ejidatarios. The society is represented in its transactions
with the Bank by a delegate. A vigilance committee is
responsible for watching over transactions between the
ejidal authorities and the Ejidal Bank and for seeing that
the credit received by the ejidatarios is used for production
purposes. The credit societies functioning in 1967 limited
their action to arranging for credit; they were not involved
in acquiring better information on crop production prac-
tices or other activities to increase net income from agricul-
tural production.
Small landowners in at least two villages in the area had
been organized in agricultural credit societies prior to 1967.
These societies consisted of landowners who wished to
obtain credit from another official bank, the National
Agricultural Credit Bank. Neither of these societies was
functioning at the time of the 1967 survey.


To obtain a better understanding of farmers' attitudes
toward organized group action and their previous experi-
ences in trying to work together, the interviews in the 1967
survey asked the farmers specific questions about matters
of organization. The majority of farmers interviewed
showed no desire to belong to an organization. Few mem-
bers of the community, even members of their own fami-
lies, were considered worthy of their trust and confidence.
Many of those interviewed declared that they preferred "to
work alone," "not to depend on anyone," and that "each
person should do whatever his means permit."

The farmers also expressed distrust of the motives of the
Puebla Project. When told that the services of the technical
assistance agents would be free, they replied that "no one
ever came to the communities with the sole purpose of
doing good." They felt that, in one way or another, they
would have to pay for the assistance. Many farmers sus-
pected that the Project was a scheme to expropriate their
land to set up an industry, or to redistribute the land. Some
farmers seemed to feel that the Project was the govern-
ment's way of finding out how much each farmer owned,
so their taxes could be raised. And there were farmers who
declared that the interviewers must be Communists,
because "only Communists go around in groups and talk
about the necessity of organizing the poor."
Clearly, in the early stages of the Project's implementa-
tion, it would have been fruitless to talk to the farmers
about the advantages of organizing groups or credit socie-
ties for participation in the action program. Thus, the new
maize recommendation in 1968. was promoted among
individual farmers on a voluntary basis.
Many of the farmers who participated in 1968 felt that
the transportation of the fertilizers was both expensive and
troublesome. If they chose to move the fertilizers by
passenger buses, the sacks were often torn and the fertilizer
spilled. If an individual farmer hired a truck to transport his
fertilizer, it was costly to haul the small amounts. Farmers
who purchased their fertilizers from local stores found the
prices to be relatively high.
These experiences caused the farmers to discuss group
action as means of lowering the costs of the fertilizers and
of transporting them to the farm. The Project's technical
assistance agents encouraged the farmers to fully explore
advantages of organized action and to discuss it with their
During the promotion stage for the 1969 growing
season, the idea of organizing groups of farmers began to be
accepted in many villages. The technical assistance agents
pointed out that, in addition to being able to transport
fertilizers at lower cost, organizing would make it easier for

the farmers to secure credit and other services. It was also
mentioned that they would be able to buy fertilizers at a
lower price than could be obtained by individuals, after the
members of a group had accumulated sufficient reserves of
The promotional activities of the technical assistance
agents in early 1969 led to the organization of 58 credit
groups with 1,556 members (Table 6.1) to receive credit
and fertilizers from a local fertilizer distributor, Agr6nomos
Unidos. The Agricultural Bank of the South participated in
the organization of 55 groups with 542 farmers, and made
credit available to them for maize production. The other
official credit banks provided credit to enable their regular
clients to use Project recommendations. As shown in Table
6.1, the National Ejidal Credit Bank authorized credit for
413 ejidatarios organized in 15 credit societies, and the
National Agricultural Credit Bank financed 50 individual
The Impulsora de Puebla, the main fertilizer distributor
in the State, participated indirectly in 1969 by providing
financing and fertilizers to Agr6nomos Unidos. The Impul-
sora continued its funding of Agr6nomos Unidos in 1970
and, in addition, provided credit directly to 253 farmers in
21 groups. In 1971, the Impulsora absorbed the clients of
Agr6nomos Unidos and continued to provide credit to a
similar number of farmers in 1972 and 1973.
The National Ejidal Credit Bank decided in 1970 to
reactivate many credit societies in the Puebla area that had
been suspended because a large proportion of their mem-
bers had failed to repay their loans. Thus, the number of
credit societies receiving credit in 1970 increased to 59,
with 2,122 farmer members. Table 6.1 shows that the
number of credit societies receiving financing from the
Ejidal Bank has remained fairly constant since 1970.
The National Agricultural Credit Bank began providing
credit to groups of farmers in 1970. It made credit available
to credit societies with 10 or more members, and to soli-
darity groups with a minimum of three members and a
maximum of nine. The solidarity group evolved as the most

TABLE 6.1. The numbers of groups and organized farmers receiving credit from several sources during the period 1969--1973
Impulsora de National Ejidal National Agricul- Bank All sources
Puebla Credit Bank tural Credit Bank of the South Others* credit
No. of No. of No. of No. of No. of No. of No. of No. of No. of No. of No. of No. of
Year groups farmers groups farmers groups farmers groups farmers groups farmers groups farmers







* The 58 groups in 1969 and 64 groups in 1970 received credit from Agr6nomos Unidos, a fertilizer distributor. The 10
groups in 1971 and 9 groups in 1972 received credit from the owner of the Hacienda Coxtocan. The 112 groups in 1973
received credit from Coxtocan and two fertilizer distributors, Agroquimica Olmeca and Guano-Mex.

attractive type of organization to the farmers, as it is very
easy to form and only one member of the group has to file
the title to his land with the bank as a guarantee. (The
members of a solidarity group accept common responsibil-
ity for debts incurred by the group.) As shown in Table 6.1,
the number of farmers receiving credit from the Agricul-
tural Bank increased to 1,865 in 1973, and the number of
groups to 314.
Since 1970, the technical assistance agents have recom-
mended the subdivision of large credit groups or societies.
In many villages, groups that receive credit from the Impul-
sora and have more than 20 members have divided to form
groups of about 10 members. Each of these subgroups has a
leader or assistant representative. Coordinating the assistant
representatives, there is a general representative or coordi-
nator who is elected by all the members and is responsible
for the organization at the village level. This subdivision of
large groups into smaller units facilitates administration,
internal communication, and contact with the technical
assistance agents.
The solidarity groups that work with the Agricultural
Bank have been encouraged to develop a similar organiza-
tion at the village level. Many of the solidarity groups have
preferred to maintain their independence, however, and this
has limited the effectiveness of their village coordinators.
Beginning in 1973, vigilance committees have been
formed in the villages with several subgroups receiving
credit from the Impulsora or the Agricultural Bank. These

From 1969 onward, the technical assistance agents concen-
trated their efforts on promoting the organization of the
farmers and in assisting the groups to use the improved
production technologies adequately. By 1973, 7,194 farm-
ers were organized in 553 groups and received credit from
seven different sources.
-.^.*r ..B

committees are composed of one member from each of the
subgroups in the village.


When the technical assistance agents began to promote
the organization of farmers in 1969, it was assumed that
organized action by farmers would be fruitful in many
ways. In 1973, selected farmers were asked whether they
were receiving benefits from working together in groups.
Information was collected from 69 farmers in 35 groups
scattered throughout the Project area. Some of the impres-
sions obtained from the farmers about the importance of
organized group action are cited in the following sections.

Better Understanding of the New Technology

The general impression of the farmers seems to be that
organization has given them greater access to information
about Project recommendations. Most of the meetings and
demonstrations at which the technical assistance agents
provide information on production practices are organized
by the groups. Members of the groups are quite consistent
in attending these sessions, and a better understanding of
the new technology by the organized farmers should lead to
a more effective use of the recommendations.
The difference between the average yields of farmers on
credit lists and all farmers in the area (Table 9.8) has grad-
ually decreases over the years. This seems to indicate that
an increasing number of farmers not on credit lists are using
the new technology. Perhaps, in terms of better informa-
tion on technology, the advantages of being organized are
most notable in the early years of the program.


Access of Small Farmers to Agricultural Credit

In practice, it is almost impossible for individual small
landholders to receive credit from an official bank or a
private institution. This is due to the high administrative
cost of a small loan, and to the fact that most small holders,
on an individual basis, cannot provide the guarantee re-
quired by the credit agency.
Farmers organized in groups receive credit from the
Impulsora de Puebla without providing any guarantee. For
the organized farmers to receive credit from the National
Agricultural Credit Bank, it is sufficient that one member
of the group deposits the title to his land with the bank. By
organizing into groups, it was possible for the number of
small farmers receiving credit in Puebla to increase from a
few hundred in 1968 to 7,194 in 1973 (Table 6.1).

Greater Efficiency in Obtaining Credit

Farmers are convinced that group action greatly expe-
dites the arranging for institutional credit. The represent-
atives of groups take care of most of the formalities re-
quired in securing credit. After the necessary papers have
been prepared by the credit agency, the group represent-
atives collect the signatures of the farmers on the individual
documents that specify the debt of each client.
The National Agricultural Credit Bank is the only agency
that requires all members of the group to go to the bank to
sign the individual documents specifying the amount of the
loan. Some groups require all members to assist in picking
up the fertilizers from the bank, or fertilizer distributor, to
reduce costs.

In the past individual small
landholders have found it
almost impossible to obtain
credit from an official bank
or private institution. Small
farmers in Puebla, by organ-
izing, have been able to ar-
range for credit. Moreover,
group representatives are able
to handle many of the re-
quirements for obtaining
loans, thus simplifying for
organized farmers the process
of arranging for credit.

In arranging for credit and picking up the fertilizers,
group representatives made an average of 5.6 trips to the
credit agency in 1973. The other members of the groups,
however, made an average of only 1.7 trips. About 47
percent of the group members did not go to the agency at
all; the representatives, assisted by the bank inspectors or
the technical assistance agents, took care of all formalities.
The costs of the trips of the representatives to the agencies
were covered by group funds.

Prompt Delivery of Fertilizers

According to the farmers interviewed in 1973, the credit
groups have been effective in reducing delays in the delivery
of fertilizers. Apparently the pressure brought by the
groups, reinforced by the technical assistance agents, has
created a new awareness on the part of the three credit
banks and the Impulsora of the importance of timely
delivery of the fertilizers. The delays in fertilizer deliveries
that have occurred in the last few years have been due to
deficiencies in the distribution of materials at the national
level, not to faulty scheduling of farmers' needs by the
agencies in Puebla.

Efficient and Cheaper Transport of Fertilizers

Another reason organized farmers have been receiving
their fertilizers on time is that the groups themselves have
made the arrangements for the transportation of the materi-
als. After the group representative receives the delivery
order, he and the other members of the group hire a truck
to haul the fertilizers at as low a cost as possible. If avail-
able, a trucker from the local village is hired for the job.

Ordinarily, a group is able to arrange for the hauling of
fertilizers for $2.40/ton. If the members of the group load
the fertilizers at the warehouse and unload them at their
houses, the cost is about $1.60/ton. Earlier, when farmers
had to arrange for transportation individually, the cost was
often as much as $4.80/ton., not including loading and

Prompt Repayment of Loans

Prior to the Puebla Project, only about 50 percent of the
short-term loans made by the official credit banks were
repaid. In 1971 and 1972, the repayment rate to the Ejidal
and Agricultural Banks was over 90 percent; the rate was
about 98 percent to the Impulsora de Puebla.
Very probably, one of the reasons for the high level of
repayment is the profitability of the new maize recom-
mendations. An even more important reason, perhaps, is
the fact that the majority of the organized farmers have
accepted a common responsibility for the debts of all mem-
bers of the group. Each farmer knows that all members of
the group must repay loans on time in order for the group
to qualify for credit the following season.
When one or more members of a group fail to repay
their loans, the rest of the group takes action to assure
repayment. As a first step, in most cases, the group repre-
sentative calls on the defaulting farmer to ask him to pay
within a fixed period. If failure to pay was due to unusual
family problems, or a poor harvest due to uncontrollable
factors, then the other members of the group may cooper-
ate to repay the debt and collect later when the farmer is
better able to make payment.
In some groups, if a member fails to pay because of
irresponsibility, the group takes firmer action to liquidate
this debt. In a few instances, the group demands some item
of property (mule, ox, plow, sewing machine, etc.) and
does not return it until the offender pays his debt. If the
defaulting member continues to refuse to pay, he is ex-
pelled from the group, and the other members repay the
loan in order to obtain credit for the following season.
One group, after trying to convince three irresponsible
members to repay their loans in 1972, took the extreme
measure of putting them in jail. Contrary to what might be
expected as a result of:this action (distrust and doubt on
the part of new members), membership in this group
climbed from 111 in 1972 to 200 in 1973. After a few days
in jail, the defaulting members repaid their loans and peti-
tioned the group to be readmitted, promising to be more
responsible in the future. The group's decision, however,
was for permanent expulsion of the three farmers.
Another reason for the high repayment of loans in
recent years is the fact that the credit agencies, with the
exception of the Ejidal Bank, have initiated the practice of
discounting interest for those who repay their loans before
they become due. Farmers with additional income during
the year prefer to pay off their loans in installments, thus
saving money they would have paid out in interest. Approx-
imately 45 percent of the farmers interviewed in 1973
liquidated their loans by making several payments through-

out the year.
Greater Efficiency in the Repayment of Loans

Many of the farmers receiving credit from the Agricul-
tural Bank or the Impulsora de Puebla make payments on
their loans directly to the general representative or coordi-
nator. This coordinator travels once a week to the agency
concerned, delivers the payments, and obtains the necessary
receipts. Thus, members not only save money on travel and
meals, but also time and effort. Generally, the expenses of
the representative are paid from a fund raised by the group
specifically for this type of activity.

Access to Information on Other Activities

According to the 35 representatives interviewed in 1973,
28 of the groups held an average of four meetings during
the year; the other seven held no meetings. The technical
assistance agents participated in about 37 percent of these
gatherings. The principal themes discussed at these meetings
were agricultural credit and the correct use of the new
maize technology. Other subjects of interest to the farmers,
such as the pruning and grafting of fruit trees, were also
discussed in some group meetings.
At harvest time in 1972, demonstrations were held in
many communities to show farmers how to construct small
trench silos for converting the maize plants into a palatable
silage after harvest of the ear but while still partly green.
The silo provides an economical means of increasing the
supply of good quality animal feed during the dry season.
Demonstrations of the pruning and grafting of fruit trees
were made in several communities. In 1971, a technical
assistance agent arranged for a group of farmers to attend a
short, course on pruning and grafting. Farmers who took
this course have been useful in assisting other farmers in the
Project area to use improved practices in the management
of their fruit trees.
Meetings and demonstrations organized by farmer
groups have often stimulated interest in a new group
activity, such as perforating a well, buying dairy cattle, or
acquiring a tractor. There seems to be an increasing aware-
ness among the farmers that the organizations should
expand their activities to include a broader spectrum of the
problems affecting the community.

Initiation of New Production Activities

At least 10 of the 385 groups functioning in 1972 were
involved during the previous 3 years in negotiating a long-
term loan for some new group activity. Most of these loans
had been requested for deep well perforations to convert a
part of the rainfed land into irrigated fields. These wells
make possible the production of higher-income crops, such
as alfalfa and vegetables, and they can be grown throughout
the year. Interest in this organized activity began to develop
after the groups were successful in using short-term credit
for maize production.
The technical assistance agents played a leading role in
this group work, from the formulation of the idea of a loan
to the reality of irrigated fields. They provided information

on the possibilities of long-term financing, helped the repre-
sentatives make contact with the credit institution that
could grant the loan, and encouraged the members of the
groups to have confidence in their capacity to work
together and in the honesty of their representatives. As the
transactions for the loans progressed, the role of the tech-
nical assistance agents tended to decline in importance,
while that of the representatives increased.
One of the groups that perforated a well and began to
produce higher income crops used the profits for a down
payment on a tractor costing $7,200. The tractor is being
used for preparing the lands of the farmers of the group,
and also for custom work for other farmers in the com-
munity, to complete payments on the tractor more quickly.
In a section of the Project area where fruit production is
important, several groups have begun to use improved tech-
nology in the management of their orchards. With the
assistance of a specialist in fruit culture, the farmers have
obtained improved varieties and transplanted them to
carefully prepared land. There is usually one man in the
group, or in the community, who has learned the proper
techniques for pruning and grafting and can teach the other

Greater Effectiveness in Solving
Community Problems

Some credit groups that have functioned for several
years and have developed relatively strong organizations
have been able to solve some of the other problems affect-
ing the community. The traditional holders of power in the
communities have come to view these groups as a threat to
their position. Local fertilizer dealers fear that the organ-
ized farmers will buy outside the community, or demand
that the dealers respect prices fixed by the national fer-
tilizer agency.
In one community, a local merchant almost went bank-
rupt in 1971 when most of the farmers, organized in
groups, began to purchase the recommended fertilizers
through one of the credit agencies. The following year, the
merchant switched from conventional fertilizers to those
recommended by the technical assistance agent in hope of
regaining his clients. Since most of the farmers were organ-
ized, however, few of them purchased fertilizers from the
merchant. (According to several representatives, the mer-
chant was prone to take advantage of temporary shortages
of fertilizers by increasing his prices, sometimes doubling
the official price.) In retaliation, the merchant began to try
to discredit the credit groups. Among other actions, he
circulated the rumor that some groups were admitting
irresponsible farmers who were unlikely to repay their
loans. When this rumor reached the credit institutions in
early 1973, some institutions decided not to accept new
clients in groups from that community.
When the farmers learned of the action of the credit
institutions, however, they solicited the help of the tech-
nical assistance agent to find a way to convince the institu-
tions to accept new clients. The group representatives and

technical assistance agent presented their complaint to the
directors of the credit institutions, but could not obtain an
alteration in policy. It appeared that the merchant's scheme
had been successful.
Nonetheless, the farmers named a commission to present
their complaint to the Governor and explain why the credit
institutions had refused to accept new clients from their
community. The commission described the standards for
admitting new members into the groups to the Governor,
showing that only responsible people were accepted. The
Governor immediately summoned the directors of the
institutions and asked them to attend to the farmers' peti-

Greater Interchange of Experiences
Among Farmers

As reported in Chapter 5, visits by groups of farmers to
other communities in the area were beneficial in many
ways. The visitors were able to directly observe the farming
activities of the group sponsoring the interchange. An
informal dialogue between visitors and hosts, with the tech-
nical assistance agent as mediator, helped members of all
groups exchange ideas and experiences about many farming
and nonfarming activities. Farmers often returned home
with new impressions about the possibilities for long-term
credit for perforating a well, or the establishment of a
small-scale dairy enterprise, fattening of pigs, pruning and
grafting of fruit trees, etc.
In 1971, a technical assistance agent in one of the princi-
pal towns in the area organized another form of exchange
of ideas among farmers. Farmers who came to town for the
weekly market day were invited, to attend an afternoon
meeting, after marketing activities had been completed. On
these occasions, about 50 farmers (mainly group represent-
atives) assembled for a lecture on a subject of current
interest to the farmers. The presentation was made by the
technical assistance agent or a specialist invited specifically
for the occasion. When the specialists spoke, the technical
assistance agent introduced the speaker and tried to clarify
any parts of the talk which he felt the farmers might not
understand. Following the presentation, there was a discus-
sion period with active participation by the farmers. This
discussion period was essential in providing the participants
with new ideas and information to communicate to their
groups on returning to their communities.


Quality of Leadership

Groups that chose a good leader at the outset have
consistently maintained good cooperation from their mem-
bers and have been able to fulfill all their obligations.
Groups with good representatives held the largest number
of meetings in 1972. Ordinarily, the representatives of these
groups sought out the technical assistance agent, or the
Project coordinator, and requested a talk on a specific

subject. Then the assistant representatives were notified of
the meeting and they, in turn, advised the members. Thus,
most members attended and maximum value was obtained
from the talk and subsequent discussion.
In contrast, the representatives who were considered
unreliable rarely held a meeting of interest to the group.
They seldom took the initiative in presenting group prob-
lems to the technical assistance agent, or in inviting him to
give a talk in the community.
Three of the 35 representatives interviewed in 1973 had
not repaid their loans on time. The members of these three
groups recognized that they had made a mistake in electing
the representatives. They pointed out, however, that it is
very difficult to select the right person. According to the
members of the groups with the irresponsible represent-
atives, these three individuals had previously had a good
record in the community; and, while the groups were being
formed, they had been very active and had given the
impression they would diligently serve the interests of the

Legalization of the Organization

Two of the 35 groups studied in 1973 had drawn up
documents outlining the regulations governing the function-
ing of the groups, and the penalties that would be imposed
on violators. These documents had been signed by the
members of the groups, the assistant representatives, the
general representative, and the municipal president, and had
been registered at the municipal headquarters. In this way,
the groups acquired legal power to take action against a
member in case he should deliberately cause problems.
Legalization of the organizations at the municipal level is
viewed by most groups as a useful step in making the
groups more efficient. In the future, it is expected that the
technical assistance agents can guide other groups in draft-
ing by-laws and in registering them with the municipal


The information collected in the study of the farmer
organizations indicates that several of the groups have made
outstanding progress in learning how to work together in
resolving problems of common interest. These groups are
now ready to move to higher forms of organization, such as
The evolution from credit groups to more complex
organizational forms will require the assistance of people
well trained in the theory and practice of farmer organ-
ization. It seems logical that the technical assistance agent
could best provide this assistance in a program such as the
Puebla Project. The present agents in Puebla, however, are
not technically prepared to do this job. To remedy this
situation, the technical assistance agents could receive
specialized training related to the operation of farmer


The experiences gained in Puebla since 1969 support the
thesis that a very simple organizational form, such as the
credit group, should be used in the initial stages of organ-
izing farmers who may often be distrustful and individual-
istic, with limited managerial ability. After the farmers have
gained confidence in the Project, and experience in col-
lective action, and have developed responsible leaders with
administrative capacity, then higher forms of organization
may be achieved. It is expected that this second stage will
be much more complex than the first, and will require tech-
nical assistance agents well trained in the organization of

Three official credit banks National Agricultural Credit
Bank, National Ejidal Credit Bank, Agricultural Bank of the
South have made loans to enable organized farmers to
use project recommendations. The total area financed by
the first two of the these banks increased from 1,516 ha. in
1969, to 13,617 ha. in 1973.



A part of the general strategy of the Puebla Project has
been to assure that the farmers have access to materials and
services essential for favorable change. Some of these serv-
ices were being provided by public and private institutions
in Puebla at the time the Project began, including: (a)
production credit, (b) agronomic inputs at a favorable
price, (c) crop insurance, and (d) an accessible market for
farm produce, with a guaranteed price.
Since these services were available in Puebla in 1967, the
Project concentrated initially on agronomic research. Had
some of these services not been available, it is unlikely that
the Project would have tried to provide them. Rather the
Project would have sought a solution through encourage-
ment of government action to create the institutions
The role of the Puebla Project in working with the
institutions (those involved in credit, the distribution of
inputs, crop insurance, and marketing) has been that of
assuring that these services are adequate for small farmers.
The Project Coordinator assumed the major responsibility
for this activity.
Planners of the Project felt it important that the Project
be promoted as a joint effort of all the agricultural agencies,
with representatives of the different institutions to be fully
informed of Project activities and the needs of the farmers.
As greater experience was gained, however, it became
evident that some of the operating procedures of some
institutions were restricting farmer use of their services. It
was clearly necessary that Project staff should more fully
understand the operations of these institutions and the way
they reached decisions on farmers' requests. Thus, the staff
began a more systematic effort to assess these institution-
al procedures.
After a problem had been fully analyzed, the Project
staff proceeded to explain the nature of the problem to the
indicated agency, usually working through the coordinator.
Full cooperation was given to the agency in finding ways to
improve its services to the farmers. This proved to be a
most difficult task, however, and progress in improving the
operating procedures of the service institutions has been
In this chapter, certain characteristics of the service
institutions in Puebla, their operating procedures, and their

accomplishments are described briefly. Changes in the
institutions that have contributed to improved services are
reported, as well as problems remaining to be resolved.


The production and distribution of chemical fertilizers
in Mexico is the responsibility of a decentralized agency of
the federal government, Guanos y Fertilizantes de M6xico
(Guanomex). When the Puebla Project began in 1967,
Guanomex had three authorized dealers in the state, Impul-
sora de Puebla and two others, all private companies. In
addition, there was a network of local distributors in the
villages, mainly retail dealers who purchased fertilizers in
the city of Puebla or Mexico City and resold them to the
Guanomex changed some of its operating procedures in
early 1971 when the three authorized distributors in Puebla
were made official commission agents of Guanomex, each
with the concession to distribute certain specific materials
at a fixed commission. Impulsora de Puebla became the
principal agent with the concession to handle low-concen-
tration materials and mixtures. According to the new
policy, local distributors in the villages would no longer be
permitted to purchase fertilizers from the official agents;
thus, they became less important in the distribution net-
Impulsora de Puebla, since its establishment many years
ago, has functioned both as a fertilizer distributor and as a
credit agency. Prior to 1968, the Impulsora had provided
credit only to commercial or semi-commercial farmers who
could offer security. In 1968, a sub-distributor of the
Impulsora, Agr6nomos Unidos, agreed to provide fertilizers
on credit to the first farmers who decided to use Project
recommendations on a part of their maize planting. These
were all small farmers and were not required to put up
collateral. Impulsora backed Agr6nomos Unidos in this
credit operation.
In 1969, Impulsora continued to back Agr6nomos Uni-
dos in its financing of small farmers who used the Puebla
Project maize recommendations. In 1970, Impulsora
continued its participation through Agr6nomos Unidos and,
in addition, provided credit directly to 253 farmers in 21
groups. Impulsora absorbed the clients of Agr6nomos Uni-
dos at the end of 1970, and has continued this credit pro-
gram for small farmers.


Table 7.1 shows the number of hectares financed, total
amount of credit, and percentage repayment of loans
within the credit operation of Impulsora, both for credit
obtained directly and that obtained through Agr6nomos
Unidos in the years 1968-1973. Impulsora's credit program
peaked in 1970 and has remained quite static since then.
This has been true in spite of a growing demand from the
farmers for more credit from Impulsora and an average
repayment rate of about 98 percent.

Procedure Followed in Granting Credit

The procedure followed by the Impulsora in granting
credit to small farmers is very attractive, since it does not
require any security from the farmers. All that the Impul-
sora requires is a guarantee of the total operation by a
responsible agency. CIMMYT provided this guarantee in
1968 when only $6,000 was loaned. Beginning in 1969, the
credit operation of the Impulsora was endorsed by the
Agricultural Agent, who is the state representative of the
Ministry of Agriculture.
In 1969 and 1970, the endorsement of the Agriculture
Agent was not backed up by adequate funds to fully
guarantee the credit operation. In 1971, however, a special
fund was formed by collecting a tax of $0.40/ton on all
fertilizers sold on credit or for cash by the three official
agents of Guanomex and the official credit banks. This
fund is administered by the State Fertilization Committee,
whose chairman is the Agricultural Agent. Since the estab-
lishment of this fund, the guarantee of the credit program
of the Impulsora has been effective. When the Fertilization
Committee reimburses the Impulsora for loans not repaid
on time, it receives the promissory notes of the indebted
farmers, with the expectation of repayment at a later date.
The Impulsora procedure for granting credit is as fol-
lows: (a) representatives of the sub-groups prepare lists of
the members desiring credit, and of the number of hectares
for which financing is requested; (b) the group coordinator

consolidates the lists of the different sub-groups and deliv-
ers the request to the Impulsora; (c) the technical assistance
agents of the Puebla Project give the Impulsora a list of all
the communities in their zones with groups requesting
credit, along with the recommended fertilizer rates; (d) the
secretaries of the Impulsora or the Puebla Project draw up
documents for each farmer, specifying the amount of ferti-
lizers and credit requested; (e) these documents are given to
the group coordinators, who, with the representatives,
obtain the signatures of the farmers; (f) the documents are
then signed by the group coordinators and returned to the
Impulsora; and (g) the delivery order is given, specifying the
date the farmers must pick up the fertilizers at the
company warehouse. In 1972, the average time required
was 36 days from the preparation of lists to the issuing of
the delivery order, with extremes of 3 days and 6 weeks.

Changes in Sales of 10-8-4

According to survey data, 64 percent of the farmers who
applied chemical fertilizers in 1967 used the 10-8-4 mix-
ture, containing 10 percent nitrogen (N), 8 percent phos-
phorus (P205), and 4 percent potassium (K20). Agronomic
research on farmers' fields, however, has not revealed
important deficiencies of potassium; thus, the Project has
recommended that farmers apply only nitrogen and phos-
phorus. Moreover, the Project has suggested that farmers
purchase nitrogenous and phosphatic materials separately
and prepare their own mixtures, to obtain the right propor-
tions of the two elements.

The relative importance of 10-8-4 in total sales of the
Impulsora declined markedly in 1969. The 10-8-4 mixture
represented approximately 76 percent of the total nitrogen
sold by the Impulsora in the period 1966-1968, whereas
only 27 percent of the nitrogen sold in the years 1969-1972
was in the form of 10-8-4. In general, there has been a shift
from 10-84 to ammonium sulfate, superhosphate, and non-
potassic mixtures, (such as 12-8-0, 10-10-0, and 5-14-0).

TABLE 7.1. The credit provided by several agencies to enable farmers to use Project recommendations for maize during the
years 1968-1973.
Impulsora de Puebla National Agricultural Credit Bank National Ejidal Credit Bank Agricultural Bank of the South Total

No, of Amount repay- No. of Amount repay. No. of Amount repay- No. of Amount repay- No. of Amount
Year hectares of credit ment hectares of credit ment hectares of credit ment hectares of credit ment hectares* of credit
1968 76 6,000 100.0 76 6,000
1969 2,719 165,059 96.0 687 48,802 50.0 829 50,846 55.5 1,603 105,132 50.0 5,838 369,839
1970 4,682 191,163 97.5 1,788 148,250 51.2 4,522 282,256 72.0 1,609 123,175 60.0 12,601 744,844
1971 3,228 108,807 99.5 4,950 202,972 91.6 4,920 294,347 91.0 1,172 90,080 62.0 14,438 696,206/
1972 4,108 153,953 98.5 7,499 383,282 93.9 5,105 398,722 90.0 822 39,854 50.0 17,533 975,811
1973 4,220 8,207 419,452 5,410 422,584 293 14,202 20,604
The total number of hectares for 1971 includes 168 with credit guaranteed by the Coxtocan Hacienda; the total for 1973 includes 2,474 ha that were financed by Guanomex
and the Olmeca fertilizer company.

The Puebla Project seems to have been largely respon-
sible for this change from 10-8-4 to more adequate fertil-
izers. Sales of ammonium sulfate and superphosphate have
increased as a direct result of the growing demand for these
materials by farmers using Project recommendations.
According to the manager of the Impulsora, his company
has given greater importance to the 10-10-0 and 12-8-0
mixtures since 1971, because the field trials conducted in
the area showed little response to potassium.

Outlook for Greater Credit for Small Farmers
From the Impulsora de Puebla

As mentioned earlier, the credit provided to small farm-
ers by the Impulsora has not increased since 1970, in spite
of a repayment rate of around 98 percent. This is due
mainly to the low interest rate which the Impulsora is
required to charge on fertilizer credit. According to the
modified Guanomex policy that became effective in 1971,
9 percent per year is the maximum interest that the Impul-
sora can charge farmers receiving fertilizers on credit. Since
the Impulsora acquires the fertilizers from Guanomex at 6
percent interest, its income from interest on credit sales is 3
percent per year. However, as credit is usually extended to
farmers for a period of about 9 months, the effective
income from interest is less than 3 percent. Under these
conditions, the tendency of the Impulsora is to increase its
cash sales and keep sales on credit to a minimum.
Perhaps permission to charge a higher interest rate is the
only measure that will induce the Impulsora to increase its

The Project coordinator
works closely with the repre-
sentatives of the agricultural
service agencies. He provides t le
information on the findings
of the Project and obstacles
limiting farmer use of the
available services, and assists

such obstacles.

sales on credit in the future. Farmers should find Impulsora
credit attractive, even with a higher interest rate, because of
the simplicity of the credit-granting procedure.


The National Agricultural Credit Bank was founded in
1926 as an integral part of the agrarian reform program of
the post-revolutionary governments. Branches of the bank
were established throughout the country to: (a) promote
the organization of ejidatarios and small landholders, (b)
make available production credit at locations accessible to
these farmers, and (c) provide for credit at an interest rate
more attractive than that charged by local moneylenders.
The mandate of the Bank stipulated that possible social
benefits be considered as well as the solvency of the client,
in deciding how to allocate its funds.
The law regulating the National Agricultural Credit Bank
was modified in 1935 with the creation of the National
Ejidal Credit Bank. Since then, the Agricultural Bank has
provided credit exclusively to landowners, and the Ejidal
Bank has worked with the ejidatarios. The law governing
the Agricultural Bank was again modified in 1956 in an
attempt to make its services more dynamic.
The National Agricultural Credit Bank has branch banks
in every state. Most branches have sub-branches or agencies
that are located at strategic points. The zone corresponding
to a given agency is divided into sub-zones, with a field
inspector in charge of each of them. The Puebla Branch of
the Agricultural Bank has six agencies, two of which pro-
vide credit to farmers in the Project area.

`hsJr 1
\ irl


The National Agricultural Credit Bank makes three types
of loans to small farmers organized into solidarity groups:
(a) short-term loans (maximum 12 months), mainly for
purchase of inputs for annual crops at 10.5 percent/year
interest; (b) intermediate-term loans (1 to 5 years) to
acquire work animals, farm machinery, dairy cattle, etc., at
9 to 10 percent/year interest; and (c) long-term loans (6 to
10 years) to purchase heavy machinery or construct farm
buildings at 7 to 10 percent/year interest.
The majority of the farmers in the Project area are not
eligible for individual credit from the Agricultural Bank,
because their holdings are too small. Although the regula-
tions of the Agricultural Bank provide for the organization
of credit societies, none of these were operating when the
Project began in 1967. As seen in Table 7.1 the Agricul-
tural Bank provided credit for farmers to use Project recom-
mendations on 687 ha in 1969 and on 1,788 ha in 1970. In
1969, credit was granted to individual farmers; in 1970, to
23 groups with a total of 480 farmers. These groups,
however, were poorly organized and there was little contact
with the technical assistance agents. Repayment was made
on only about 50 percent of the loans made in these two
In 1971, the Agricultural Bank and the Project technical
assistance agents began to promote the organization of
solidarity groups. These groups have a maximum member-
ship of nine and a minimum of three. According to bank
regulations each member of these groups must mortgage his
land to the bank in order to obtain credit. As most small
farmers do not have a clear title to their land, the Puebla
Branch of the Agricultural Bank obtained authorization to
change this requirement. Now, the only requirement is that
one member of the group has a clear title to his land and is
willing to mortgage it to the bank. This change in regula-
tions has made it possible for many farmers who were
previously unable to qualify for loans to obtain credit as
members of a solidarity group.

-~Arie ; ...:>..rP

Organized farmers are encouraged to repay their loans as
soon as possible. Many farmers liquidate their loans by
making several payments throughout the year. In 1972, the
percentage repayment on loans from the Impulsora de Pue-
bla was 98 %, from the National Agricultural Credit Bank
94% and from the National Ejidal Credit Bank 90% .

With the change to solidarity groups in 1971, the Agri-
cultural Bank has greatly increased its credit to farmers
using Project recommendations. Table 7.1, shows that the
Agricultural Bank provided credit for 4,950 ha in 1971;
7,499 ha in 1972; and 8,207 ha in 1973. Repayment of
loans by solidarity groups exceeded 90 percent during this

Operating Procedures of the Agricultural Bank

Farmers in solidarity groups follow these steps in arrang-
ing for credit with the Agricultural Bank: (a) the group
representative prepares a list of the members that indicates
their ages, beneficiaries in the case of death, and the areas
for which credit is requested; (b) one farmer with a clear
title to his property agrees to guarantee the group loan-
both he and his wife must register their signatures with the
legal department of the bank; (c) each member presents the
receipt for his most recent property tax payment, or a
letter from the highest authority in the community certify-
ing that he is a property owner; (d) the representative
obtains a written statement from the technical assistance
agent indicating the fertilizer rates recommended for the
group; (e) a credit application is drawn up for the group,
specifying the total area for which credit is requested and
the total amount of credit-this application is signed by the
group representative, by the guarantor, and by the technical
assistance agent; (f) based on the application, a special
form, called F-200, is prepared as a credit application from
the group to the Planning Council of the bank-these
F-200's have to be signed by five persons in the Credit,
Legal, and Administrative Departments, as well as by the
bank manager; (g) the group signs a contract with the bank
for the amount of credit requested-the contract must be
registered in the city of Puebla with the payment of a
registration fee of $8.00; the contract is good for 5 years,
provided there are no changes in the group; (h) all members
sign a formal charter, specifying that they assume a com-
mon responsibility for the obligations of the group--this
means that should any member fail to repay his loan, the
others are obligated to find a way to liquidate it; (i) each
member signs a letter in which he agrees to repay his loan
with the produce from the land for which credit was
received; (j) multiple promissory notes are drafted showing
the cost of fertilizers, amount of interest, and other charges
for each member-these documents are prepared with 18
copies and signed by all members of the group; and, finally
(k) the guarantor deposits a letter with the bank giving it
power to take possession of his property in case the group
fails to meet its obligations.
In 1972, this procedure required an average of 13 days
from the time the application was presented until the group
received the order to pick up the fertilizers. Some groups
were able to complete this process in 3 days; others re-
quired as much as 4 weeks.
Some farmers felt this procedure for arranging for credit
was too complicated and chose not to join a solidarity
group. Other farmers were discouraged from seeking credit


through the Agricultural Bank because of the requirement
that they take crop insurance. (This latter problem is
common for the three official banks and is discussed in a
later section.)
When the number of farmers seeking credit from the
Agricultural Bank increased sharply in 1971, the bank ran
short of personnel to handle the loan requests. The farmers
in solidarity groups suggested that the Bank use the 1
percent of their loans that is deducted by law for admin-
istrative costs to pay temporary personnel. The director of
the Bank presented the proposal to the local administrative
council and obtained the necessary approval.

Outlook for Greater Credit for Small Farmers
From the Agricultural Bank

The Agricultural Bank has become a strong supporter of
the Project strategy and has fostered close coordination of
Bank and Project personnel in their relationships with the
farmers. Both the Puebla Branch and the Central Office of
the Agricultural Bank are pleased with their credit program
for solidarity groups using Project recommendations. All
indications are that the Agricultural Bank will seek to
increase this credit operation. Prospects for this increase
were strengthened in 1973 when the Agricultural Bank
received authorization to extend credit to ejidatarios as well
as landowners.
Nevertheless, if the Agricultural Bank is to respond fully
to the growing demand for production credit in the Puebla
area, it will be necessary to resolve several problems that
presently limit the effectiveness of the bank's credit pro-
gram: (a) the work load of employees of the Legal and
Administrative Departments has increased greatly since
1971-it seems reasonable their salaries should be adjusted
to compensate for this work, and to put their salaries in line
with those received by employees of the other official
banks; (b) bank regulations require that a new contract be
signed with a solidarity group when members leave, when
new members enter, or when there is a change in the area
for which credit is requested. Since such changes are
frequent, it is necessary to sign a new contract almost every
year. Farmers feel that the bank should be more flexible
and permit changes in membership and acreage without the
trouble and expense of drawing up and registering a new
contract; (c) the Agricultural Bank is the only official bank
that deducts interest in advance. A farmer who requests
$100 at 11.5 percent interest, for example, receives $88.50
and has to pay back $100 at the end of the year. The
effective interest rate is 12.99 percent instead of 11.5
percent. The farmers feel that this method of calculating
interest should be changed; and (d) the temporary person-
nel hired for the peak work periods are usually poorly paid
and inefficient. Thus, many documents have to be redone,
causing delays and a great deal of frustration for the farm-


The National Ejidal Credit Bank was founded in 1935
and since that time has granted credit to ejidatarios, with
the Agricultural Bank providing credit to landowners. In
addition to extending credit to ejidatarios, the Ejidal Bank
objectives were: to organize the ejidatarios to work their
lands collectively, and to make credit more accessible by
establishing state and regional branches. The Puebla Branch
of the Ejidal Bank has seven agencies at strategic locations
in the state. Each agency has several field inspectors, each
of whom is responsible for attending the ejidatarios in a
given zone. Two agencies are located in the Project area and
have provided credit to ejidatarios using Project recom-
The Ejidal Bank, as in the case of the Agricultural Bank,
offers short, intermediate, and long-term loans. The Ejidal
Bank makes loans to credit societies and, rarely, to solidar-
ity groups. Credit is not provided to individual ejidatarios.
In 1969, the Ejidal Bank granted loans enabling 413
ejidatarios in 15 credit societies to use Project recom-
mendations on 829 ha of maize. Although only 55 percent
of these loans were repaid (Table 7.1), the Ejidal Bank
recognized the potential of the new technology and decided
to reinstate 44 ejidal societies that had lost their credit
standing in the past due to failure of their members to
repay loans. In 1970, the Bank provided credit for 2,122
ejidatarios in 59 credit societies with a total of 4,522 ha.
Table 7.1 shows that the number of hectares of maize
plantings financed by the Ejidal Bank has remained fairly
constant since 1970. About 90 percent of these loans have
been repaid.

Operating Procedures of the Ejidal Bank

To qualify for loans from the Ejidal Bank, the ejidatarios
must organize a credit society. The requirements for char-
tering a society are: (a) at least 10 ejidatarios must request
the founding of a society; (b) the ejidal executive commit-
tee must submit an application for the formation of a
society, along with a map of the ejido showing the location
of the parcels of the applicants; (c) each applicant must
present his ejidal certificate, or proof that his name appears
on the most recent ejidal census list-this is necessary to
establish that each of the applicants has possession of an
ejidal parcel; (d) the field inspector of the Ejidal Bank
prepares a document specifying the socioeconomic condi-
tions of the applicants and the productive potential of their
soils; (e) the Puebla Branch of the Ejidal Bank sends the
application of the ejidatarios, the map of the ejido, and the
document prepared by the field inspector to the Central
Office; and (f) the Central Office approves or disapproves
the application.

There is at least a 6-month lapse from the time the
application for the formation of a credit society is sub-
mitted until the Central Office makes a decision. The Bank
does provide, however, for the provisional establishment of
a society under unusual circumstances and at the request of
the manager of the Branch Bank.
After the formation of a credit society has been author-
ized, the field inspector holds a meeting, in the efido, of all
the ejidatarios who wish to become members and receive
credit from the Ejidal Bank. At this meeting, the ejidatarios
elect one member of their group, the delegate, to represent
the society in all transactions involved in arranging for
credit. Once established, the credit society can function
indefinitely. New members can be accepted on approval of
the membership of the society. The Ejidal Bank is fairly
flexible in its relationships with the society and permits it
to continue operating even though there are changes in its

The procedure followed by the credit society in arrang-
ing for loans from the Ejidal Bank involves several
steps: (a) during September preceding the cropping sea-
son, the field inspector, with the assistance of the delegate,
prepares a list of the credit requirements of each member;
(b) the Puebla Branch consolidates the requests from the
several societies, prepares a Plan of Operations, and sends it
to the Central Office; (c) the Central Office approves all or
part of the request and returns it to the Branch Bank; at
least one month before planting time, the Branch Bank noti-
fies the agencies of the amount of credit approved for their
areas; (d) the field inspector prepares a final list for each
society, showing the credit requested by each member (last-
minute changes in members requesting credit is permis-
sible); (e) the field inspector, assisted by administrative
personnel of the Bank, prepares a contract that specifies the
amount of credit requested by the society, both for fer-
tilizers and in cash. A new contract is necessary each year
for each type of credit. The ejidatarios, however, do not
participate in the preparation and registration of the con-
tracts. This is done by the Bank, which pays the registration
fee from a special fund collected from society members for
administrative expenses; (f) the field inspector draws up a
multiple promissory note showing the amount of each
member's loan, the interest, and other deductions; (g) the
Bank then sends a delivery order to the delegate-with this
order, the society members obtain their fertilizers and cash
from the Bank; and (h) each member, on receipt of materi-
als or cash, signs the promissory note. In 1972, the average
time between the preparation of the final list (step d) and
receipt of the delivery order was 28 days.
In an attempt to make the credit-granting process more
dynamic, the National Ejidal Credit Bank introduced a
series of administrative reforms in 1973. The central feature
of these reforms is the creation of Control Boards with an
assistant bookkeeper and secretaries, to relieve field inspec-
tors of most of their paper work. Its is hoped that this will
enable the field inspectors to devote more time to field

In early 1973, by presidential decree it was stipulated
that debts contracted by ejidal credit societies between
1940 and 1965 would be pardoned at the rate of 20 per-
cent for each year that the efidatarios repay new loans on
time. Thus, in 5 years, all ejidatarios with debts from that
period could liquidate the old loans simply by repaying all
new loans promptly. The presidential decree further spe-
cified that debts contracted between 1966 and 1972 would
be combined into one account and repaid gradually without
interest. The Bank suggested that each indebted ejidatario
apply 10 percent of his harvest each year toward paying off
his account until it was settled. This new policy for reinstat-
ing members of the credit societies should increase the
amount of credit granted in 1974 to enable ejidatarios to
use Project recommendations.

Outlook for Greater Credit for Ejidatarios From
the Ejidal Bank

Taking into account the 90-percent rate of loan repay-
ment since 1971 by ejidatarios using Project recommenda-
tions, it seems likely that the Ejidal Bank will be interested
in expanding its credit operation in the Puebla area. To
accomplish this, however, the Bank must examine a series
of problems that are believed to limit the effectiveness of
its credit program, including: (a) eight field inspectors
attended 52 societies with 2,499 ejidatarios in 1972, an
average of 6.5 societies and 312.4 ejidatarios per field
inspector. This is felt to be about the maximum number of
societies and members that can be attended efficiently by
one inspector with present operating procedures. It seems
clear, therefore, that more inspectors will be needed or
operating procedures must be simplified if the Ejidal Bank
is to expand its credit operation successfully; (b) In 1969
and 1970, technical assistance agents of the Project focused
their efforts on groups receiving credit from the Impulsora
de Puebla. Since then, they have tried to work more closely
with the ejidal credit societies. In 1973, for the first time,
technical assistance agents and bank inspectors began to
hold meetings to coordinate their activities. Both groups
would benefit from strengthening this relationship in the
future to improve their services to the ejidatarios; (c) when
communicating with Bank administrators, the field inspec-
tors sometimes fail to transmit the nature of problems
influencing the behavior of the ejidatarios. This lack of
communication results in misunderstandings and strained
relationships among Bank employees. Recently, for exam-
ple, field inspectors were blamed for the failure of many
ejidal parcels to qualify for crop insurance. However, there
was little the inspectors could do, because the problem
arose primarily as a result of unrealistic operating polices of
the crop insurance agency, plus the tendency of some farm-
ers to encourage rejection (believing the cost of the insur-
ance to be a useless expense). Regular meetings of field
inspectors and other bank personnel would permit a freer
flow of information and contribute to a smoother function-
ing of the institution; (d) in past years there have been an
excessive number of changes in administrative personnel

and field inspectors. In one year, for example, the Bank
manager was changed four times. Frequent changes in Bank
personnel cause many deficiencies in the bank's services to
the ejidatarios; and (e) members of the ejidal credit soci-
eties are poorly informed of their rights and obligations and
few participate in transactions with the Bank. Most mem-
bers, for example, do not know what percentage of interest
they are paying, how much is deducted from their loans for
crop insurance, or why other deductions are made. Society
members are poorly informed mainly because they fail to
attend the meetings held by the field inspector. According
to the ejidatarios, however, nothing of importance is discus-
sed at the meetings. This situation can be improved in the
future if the field inspector will devote more time to vis-
iting the ejidos, perhaps soliciting the participation of the
technical assistance agents.


The National Crop and Animal Production Bank was
established in 1965, with the function of granting credit to
both ejidatarios and landowners. Its basic purposes are to
assist the Agricultural and Ejidal Banks in serving more
farmers, and to seek new ways to make these services more
dynamic and efficient.
The National Crop and Animal Production Bank has
four regional banks that function independently in admin-
istrative matters, each providing service in several states.
One of these regional banks, the Agricultural Bank of the
South, was established in the city of Puebla in 1967 and
serves eight southeastern states, with agencies in each.
The operating procedures of the Agricultural Bank of
the South are similar to those of the Ejidal Bank, the dif-
ference being that farmers can organize solidarity groups (if
their membership is less than 10) or credit societies (if
membership is 10 or more). The time required to found a
society is about the same as in the case of the Ejidal Bank.
An important difference between the Agricultural Bank of
the South and Ejidal Bank is that the Ejidal Bank field
inspectors collect payments on loans directly in the com-
munities and, when necessary, from the ejidatarios in their
homes. The field inspectors of the Agricultural Bank of the
South do not collect loan payments.
When the Puebla Project began to promote the use of
the new maize recommendations, the Project technicians
and farmers felt that the potential of the Agricultural Bank
of the South for supplying production credit exceeded that
of the other official banks. It was new, well-equipped, had
well-trained personnel, and was interested in new ap-
Table 7.1 shows that the credit provided by the Agricul-
tural Bank of the South peaked in 1969 and 1970 and
steadily declined in the following three years. This tenden-
cy of the bank to reduce its credit program apparently
stems from two causes: (a) the percent repayment on loans
has been low, only 50 to 62 percent-this low repayment
rate can be explained in part by the fact that the Bank's
clients have received little technical assistance due to the

lack of coordination between the Bank's field technicians
and Project personnel; and (b) the Bank does not feel that
short-term credit, especially for maize production, is an
effective way of helping small farmers. The Bank feels that
such loans tend to perpetuate the vicious cycle of poverty-
subsistence that is at the root of underdevelopment. In
extending credit to a considerable number of farmers in
1969 and 1970, this Bank sought to introduce them to the
Bank's services with the aim of promoting long-term loans
to make the farmer's operations more productive. Few of
the farmers, however, reacted as the Bank had expected.
Since 1972, the Agricultural Bank of the South has
promoted a different form of organization, which consists
primarily of consolidating the contiguous holdings of a
group of farmers and operating the land as a single unit.
The bank expects to drill wells and convert most of the
land to higher-income, irrigated crops. This organizational
model is presently being tried with three societies in the
State of Puebla.
In view of the present thinking of the Agricultural Bank
of the South, it seems unlikely that the Bank will grant
significant amounts of credit in future years to enable farm-
ers to use Project recommendations.


In 1969, the owner of the Coxtocan Hacienda gave
chemical fertilizers to many of the ejidatarios who farmed
the land adjacent to her property. The following year she
was deluged with requests for fertilizers and was forced to
look for some other way to assist the ejidatarios. 'The
coordinator of the Puebla Project suggested that, rather
than give the fertilizers, the ejidatarios might request the
fertilizer on credit from a distributor, and that she could
serve as guarantor for the loans. The Project agreed to
organize the ejidatarios and provide them with technical
Since 1971, the owner of the Coxtocan Hacienda has
guaranteed the loans for about 250 ejidatarios in 10 groups.
The credit was extended by the Olmeca fertilizer company
in 1971 and by the Impulsora de Puebla in 1972 and 1973.
It is not expected that the owner of the Coxtocan Hacienda
will be equipped to guarantee the loans of larger numbers
of ejidatarios in future years.


As mentioned earlier, Guanomex is a decentralized
federal agency with the responsibility for the production
and distribution of chemical fertilizers in Mexico. It ini-
tiated a pilot effort in 1973 to promote the use of agua
ammonia by organized farmers in the Puebla area. Guano-
mex feels that nitrogen in the form of agua ammonia can be
supplied to small farmers at a cost of only about 60 percent
of that of solid fertilizers. An important factor contributing
to this lower cost of liquid fertilizers is the feasibility of
transferring the personnel and special equipment for hand-
ling agua ammonia presently assigned to irrigated areas of
the country to rainfed areas for a few months each year.

The Guanomex plan was to provide farmers with horse-
drawn applicators and deliver the agua ammonia to their
communities in 55-kg tanks that fit directly on the appli-
cators. The farmers had to sign a promissory note on
receipt of the fertilizer and agree to repay Guanomex at
harvest time.
The Project technical assistance agents began to inform
the farmers about the Guanomex program in early April
1973. Over 2,000 farmers volunteered to try the agua
ammonia on some 5,000 ha. Several problems arose, how-
ever, including delays in installing the agua ammonia plant,
and numerous difficulties in adapting the horse-drawn
applicator to the conditions in Puebla. Thus, the agua
ammonia was applied to only about 500 ha by about 250
Project technicians feel that agua ammonia can be an
important source of nitrogen for farmers in the Puebla area,
if the price can be maintained at around 60 percent of that
of solid materials. However, the horse-drawn applicator
used in 1973 still has many technical flaws and will have to
be improved significantly.


About 20 years ago, farmers of the Lagunera Region of
Northern Mexico who received credit from the official
banks formed a mutual- crop insurance association. Each
member paid a fixed amount per hectare, and in the case of
crop damage, the money was distributed among those mem-
bers suffering losses in accordance with the recommenda-
tions of an inspection committee named by the association.
This mutual association was quite successful, and similar
agencies were soon formed in other parts of the country.
The first mutual association in the State of Puebla was
established in 1956 and became a part of the National
Agricultural Insurance Agency (ANAGSA) in 1961.

The basic purpose of ANAGSA is to complement the
agricultural credit service provided by the official banks, by
protecting: (a) the farmers against losses due to natural
causes, and (b) the official banks against losses due to the
inability of the farmers to repay their loans in unfavorable
years. ANAGSA expanded its program in 1972 to include
life insurance, which costs the farmer $2.00/year and
provides his family with $400 indemnization in case of
Table 7.2 shows the hectares of maize insured by
ANAGSA in the Puebla area, the premiums paid, and the
amounts of indemnizations for the years 1966-1971.
Although all farmers applying for official production credit
must request crop insurance (except for one line of credit
of the Agricultural Bank), ANAGSA normally rejects some
of the plantings. For example, in 1971, only about two-
thirds of the area receiving credit for maize production was
approved for crop insurance.
The average area of maize insured by ANAGSA in
1970-1972 was more than double that of 1966-1969 (Table
7.2). The premiums paid by the farmers accounted for less
than one-third of the total premiums; the remainder was
paid by the federal government. The area on which indem-
nization was collected has fluctuated greatly, reflecting
variations in climatic conditions over the years.

Operating Procedures of ANAGSA

The procedure used by the official credit banks in
requesting crop insurance for their clients is as follows: (a)
each bank sends a multiple application to the insurance
agency with the areas for which credit is requested and
names of all farmers soliciting credit, and (b) as soon as the
farmers have signed the contract and promissory notes, the
Bank sends a complementary report for each farmer to the
insurance agency, showing his age, beneficiary, and the
number of parcels in his property that are separated by

TABLE 7.2. The insuring of maize plantings in the Puebla area in 1966-1972.

Area financed Area Amount Premiums Area
by official banks insured of paid by Indeminized Amount of
Year ha ha premiums farmers ha indemnization

1966 2,973 2723 42,832 271 4,421
1967 3,187 2740 45,161 1185 23,069
1968 3,545 2856 45,405 -- 710 10,130
1969 3,118 2672 45,048 12,456 1840 51,079
1970 7,920 6139 103,408 /29,598 1186 34,762
1971 11,043 7068 176,538 45,914 1532 59,304
1972 13,426 5947

more than 1 km. The insurance agency uses the information
in these reports to estimate the number of field inspectors
needed and the approximate dates of peak field activities.
Three kinds of field inspections are made. The first
inspection is normally made a few weeks after emergence of
the crop. Its purpose is to verify that the insured crop was
planted, and that plant density and general vigor are sat-
isfactory. The group representative and the members
needed to locate the parcels must accompany the field
inspector. After each parcel is checked, the inspector draws
up a legal document specifying the conditions of the crop.
The document is read to all those present and signed by
each. The inspector does not inform the farmers at this
time whether the parcel is accepted or rejected; this
decision is made later by a higher agency official.
The second kind of inspection is made when insured
farmers report crop damage due to natural causes. The
farmer must advise the insurance agency within 72 hours of
the time the damage occurs. The agency must inspect the
damaged crop within a period of time specified by the law
governing ANAGSA. The damaged crop is inspected by the
agency field inspector, accompanied by the bank inspector,
group representative, and the farmer concerned. The inspec-
tor draws up a legal document specifying the nature and
extent of damage and has it signed by all present.
The third kind of inspection is made just before harvest
All plantings that have been reported as being damaged
during the year, are inspected to determine how much they
should produce. The field inspector reports this informa-
tion to a higher agency official who decides on indem-
Up until 1973, the inspection at harvest time was made
on all parcels belonging to the farmer reporting damage, not
just the damaged' parcel. This was because the insurance
agency did not consider the insured unit as a single parcel,
but as all parcels of the farmer concerned. Field inspectors,
therefore, estimated the yields of all the parcels and took
an average. If this yield was greater than the limit below
which indemnization was paid, the farmer received no
compensation, even though one of the parcels were a total
loss. This procedure for determining indemnization has
been the principal source of dissatisfaction with the service
of the insurance agency among the farmers.
Aware of the farmers' attitude toward the insurance
,agency's procedure for approving indemnization, the Puebla
Project staff consulted with the agency's director in 1972
about ways to resolve the problem. The insurance agency
agreed to treat parcels separated by more than 1 km as
separate insured units. This modified policy went into
effect in 1973.

Crop Insurance Related Problems That Limit
Farmer Use of Official Credit

Crop insurance in Puebla is seen as a major factor limit-
ing farmer use of official credit. Although information
collected in surveys indicates that farmers feel crop insur-

ance is necessary in the region, the majority of them would
not use it under present circumstances, if it were optional.
As the farmers see it, the insurance agency is protecting the
banks against losses, but not themselves.
To gain the confidence of the farmers in Puebla, changes
must be made in the insurance agency's operating proce-
dures. In addition, closer cooperation is needed with the
credit banks and the Puebla Project, in coordinating effec-
tively the field activities of agency inspectors, bank inspec-
tors, and technical assistance agents. Several activities that
could increase the effectiveness of the insurance agency are:
(a) informing the farmers of the crop insurance law: Most
farmers have little or no knowledge of the crop insur-
ance law, even about essentials such as the time period
for advising the insurance agency in case of crop dam-
age. However, field inspectors of the agency and banks,
together with Project technicians, could provide in-
structions about the role of crop insurance, rights and
obligations of the insured, and other operating proce-
(b) defining the criteria for rejecting plantings because of
"imminent risks": At the present time the insurance
agency rejects plantings because of "imminent risks."
Apparently, however, the reasons for rejecting such
plantings are not well-defined. This creates dissatisfac-
tions among farmers and can be avoided by precisely
defining the nature of "imminent risks."
(c) informing the farmers promptly about the acceptance
or rejection of their plantings and their right tb indem-
nization. After the field inspectors look over a planting
and prepare a report, farmers usually conclude that the
planting is insured. This may or may not be the case,
however, since the decision to accept or reject a plant-
ing is made in the Puebla office of the agency. Sim-
ilarly, in the case of inspections at harvest time, farm-
ers may erroneously conclude that their claim has been
accepted when the inspector makes no statement to
the contrary. It is important that farmers fully under-
stand how decisions are made on these matters and that
they be advised within a few days following inspec-
tions of the action taken by the insurance agency.
(d) developing a field inspection procedure that is more
efficient for a highly fragmented area such as Puebla.
Farmers in Puebla have an average of about 3 culti-
vated hectares distributed among three to four parcels
that are often separated by a kilometer or more. Many
of these parcels cannot be reached by vehicle; thus, the
process of inspecting parcels is very laborious and
expensive, resulting in long delays in making inspec-
tions and more problems for the farmer. Because of
delays in inspections at harvest time, for example, the
farmer may be required to postpone cutting and shock-
ing his maize and plowing the land. As a result of such
delays, farmers may not be able to plant early with
residual moisture the following spring.
A procedure similar to that used by the original
mutual crop insurance associations might be a possible
solution to this problem. Each group would name a

committee to inspect members' parcels and render a
report. The agency field inspectors would make spot
checks periodically to assure that the committees were
observing the regulations of the insurance agency. A
procedure similar to this is presently being used with
good results by one group that receives credit from the
Impulsora de Puebla.
(e) to cooperate more closely with the credit banks and
the Puebla Project. The bank managers and the Project
coordinator saw their relations with the director of the
crop insurance agency gradually deteriorate during
1972 and 1973. The primary reason for this weakened
relationship seems to be the questions asked by the
Project staff regarding the operating procedures of the
insurance'agency. Near the end of 1972, for example,
the Project coordinator invited the bank managers and
the insurance agency director to work with him in
finding a way to modify certain procedures of the
insurance agency that were very troublesome to the
farmers. The director of the insurance agency, how-
ever, showed little inclination to cooperate. The only
accomplishment of the meeting was the decision to
consider parcels separated by more than 1 km as dif-
ferent units for insurance purposes. Increased coopera-
tion between the agency director and the represent-
atives of other agricultural institutions in Puebla is
essential to a more effective crop insurance program.


The National Marketing Agency (CONASUPO) was
established in 1962 with the following objectives: (a) to
improve rural family income through support prices for
different agricultural products; (b) to maintain reserves of
basic foods with which to meet possible shortages; and (c)
to regulate prices in consumer markets of basic foods to
protect the low-income population.
Until 1971, CONASUPO purchased maize in the Puebla
area and stored it in the central warehouses of the National
Storage Agency (ANDSA). During 1968-1970, however,
CONASUPO constructed rural warehouses at 15 locations
throughout the Project area. Since 1971, CONASUPO has
purchased maize at these warehouses without restrictions as
to grain color or minimum quantity per producer. It has
paid the official price of $75.20/ton for grain containing 14
percent moisture or less. Prior to the harvest in 1973, the
official price was increased to $96/ton.

The purchasing procedure of CONASUPO is relatively
simple. An employee receives the maize from the producer
at the warehouse, weighs it, and determines the moisture
content. If the grain contains more than 14 percent mois-
ture, its weight is adjusted to that moisture content. A sales
slip is prepared showing the weight of grain received and its
value. The farmer presents the sales slip to the cashier and
receives his payment.
One problem with this procedure, as far as farmers are
concerned, is that the cashier is normally present at the
warehouse only 2 days per week. This means that farmers
often have to make a second trip to the warehouse in order
to get paid. For this reason, and the inconvenience of
having to haul their grain to the warehouse, most farmers
continue to sell their maize to the village grain merchant.
The network of CONASUPO buyers, nonetheless, serves to
keep the price paid by the local buyer from falling much
below the official price.
Table 7.3 shows the amounts of maize purchased in
1971-1972 and 1972-1973 at the 15 rural warehouses, as
well as the amounts sold to local consumers. Purchases in
these 2 years account for a small part of the total maize
sold in the area, indicating that most farmers sold to local
buyers. The amounts purchased in 1972-1973 were much
lower than in 1971-1972, probably because the price
offered by local buyers that year was above $75.20/ton,
due to a general shortage of maize throughout the country.

TABLE 7.3. Metric tons of maize purchased and sold by
Conasupo at the rural warehouses in the Puebla area.
Name of Purchases Sales
warehouse 1971-72 1972-73 1971-72 1972-73

Magdalena 1463 675 7.4 18.7
Ocotitlan 141 45
Tlaxco 6485 3606 14.9 74.9
Coatepec 742 192 56.6
Malacatepec 191 105 96.8
Benito JuArez 1077 343 67.7
Teotlalcingo 281 283 84.6 0.9
Tlautla 2660 1932 13.1
El Verde 751 73 4.5 34.3
Acozautla 590 34 2.6
Tepeaca 3643 1278 210.7
San Andr6s Cholula 1534
Emiliano Zapata -- 460 29.1
Guadalupe Zaragoza 20 4.4
Santiago Coltzingo 7 -

Total 19,558 9,053 111.4 609.8



The Puebla Project was conceived as an experimental
approach to develop and test strategies for rapidly increas-
ing yields on small land holdings of subsistence farmers. Its
operational strategies were designed to be flexible and
subject to modification as new information was generated.
Thus, provision was made for an evaluation unit with two
main objectives: (a) to measure the progress made by the
Project over time, and (b) to identify obstacles and collect
the information needed for modifying strategies. Immediate
feedback of this information to Project staff was a crucial
step in developing remedial actions.
Consideration was given to contracting an independent
agency to make the evaluation. There were two principal
arguments in favor of this approach: (a) greater objectivity
could be expected, as those involved in evaluation would
have no direct personal interest in the success or failure of
the Project, and (b) these more objective findings would
probably carry more weight with policy makers.
There were also two important reasons for including
evaluation as an integral part of the Project: (a) it would
assure a continuous feedback of information to the other
members of the Project team, and (b) obstacles limiting
farmer participation could be identified and studied most
effectively by an evaluation team working side by side with
the members of the field staff.
After discussion of alternatives by Project members and
advisors, evaluation was included as an integral part of the
Project. In regard to the question of objectivity, it was felt
that the essential conditions were objective criteria and
adequate methodology, as in any research.
After the harvest of the fertilizer experiments in 1967, it
became evident that the Project would begin promoting the
use of revised maize technology in 1968. Thus, it was
necessary to begin immediately to: (a) establish bench-
marks on yield, technology of the farmers, level of living,
etc. for future comparisons; (b) obtain information about
the farmers and their present level of technology for use in
planning the action program; and (c) obtain information on
the infrastructure of the region-fertilizer distribution,
agricultural credit, crop insurance, and price support pro-


Unpublished data for 1960 were obtained from the
Census Bureau for the municipios in the Project area. These

data provided a general idea of the area, the number of
families living there, the total area planted to maize and the
amount produced, and the size of the holdings.
Yearly data on area, production, and yield were available
by municipios from the Direcci6n General de Economia
Agricola. The methods of data collection and yield estima-
tion used by this agency were studied to determine whether
such data would provide an adequate estimation of yield
changes. This analysis suggested that a more precise meas-
ure of yield was needed to detect minor year-to-year
To obtain the necessary kinds of estimates of both yield
and characteristics of the farming population, a probability
sample was chosen. This sample was used both for personal
interview surveys and for yearly estimates of maize yields.


Survey: 1967

Farm operators of the Project area were the population
of interest in this study. Because the Census lists were 8
years old, it seemed advisable to look for an alternative
sampling frome. An area sampling technique turned out to
be feasible because of the availability of aerial photos taken
just 6 months earlier.
To keep costs at a reasonable level, a two-stage sample
was drawn. The sample was selected as follows. Using a map
of the region provided by the Mexican Defense Depart-
ment, the Project area or was delineated. Next, 25 points
were identified by locating coordinates with a list of ran-
dom numbers. These points were then transferred to the
aerial photos and a square 5 x 5 cm was drawn with the
point-as the center. This 25 cm2 area was equal to 100 ha.
These squares were then photographed and enlarged to a
size which simplified identification of individual parcels and
permitted more precise measurements of area.
The first stage of the field work involved locating the 25
segments and finding reference points-trees, roads, gulleys-
that would help to identify individual parcels. Once the
segment boundaries were established, the next step was to
obtain the names of those who had operated each piece of
land in 1967. The list of names of persons farming any land
within the segment constituted the sampling frame for the
second stage.

Random samples of the farm-
ers in the area were inter-
viewed in early 1968 and in
mid-1971. The information
collected in these surveys was
used to describe the condi-
tions existing at the beginning
and to measure the changes
that had occurred after the
Project had been operating
for three years.

The number of segments to be included in the sample
and the number of farmers needed in each segment were
estimated from the variability in two sets of data: (a)
yields from the fertilizer trials planted throughout the area
in 1967, and (b) yields measured on a sampling of farmers'
fields in two municipios of the area in the fall of 1967.
Based on these data, a 12 percent random sample was
drawn among farmers in each segment to assure that 10
percent would be scheduled for interviews. A total of 251
farm operators were interviewed in the 25 segments.
Interviewers used a questionnaire that was pretested in
December 1967. The questionnaire was designed to obtain
information on: (a) types of farm ownership, (b) crop
production, (c) livestock production, (d) farming costs, (e)
tools and equipment, (f) composition of family income and
its distribution, (g) crop production information at the
farmers' disposal, (h) knowledge and use of modern agricul-
tural inputs, (i) marketing of agricultural products, (j) exist-
ing organizations, (k) demographic and cultural character-
istics of the farmer and his family, (1) living conditions of
the farmer and his family, and (m) attitudes and expecta-
tions of the farmers about the future of agriculture.
After the 1967 pretest, the questionnaire was revised.
Students, principally from the National School of Agricul-
ture at Chapingo, conducted the bulk of the interviews
during the 6-week period from January 2 to February 15,
The questionnaires were checked, and the information was
coded and then punched on computer cards. The cards
were computer processed, using specially designed programs
to produce condensed tables. The card punching, pro-
gramming, and processing were done at the Statistics and
Computation Center of the Graduate College, Chapingo.

Survey: 1970

A second personal interview survey was conducted in the
summer of 1971, referred to in this report as the 1970
survey because the data corresponded to events of 1970.
The main objectives of this study were: (a) to quantify
the changes occurring between 1967 and 1970, (b) to
examine factors that might have influenced these changes,
and (c) to identify the factors that were favorable or
unfavorable to the progress of the Project. Because com-
parisons had to be made over time, the questionnaire
included the questions used in 1967, plus questions about
variables not previously studied. The information collected
in 1971 was designed to serve as a new benchmark for
future studies.
Two categories of farmers were delineated in the Puebla
area in 1971: (a) farmers on credit lists who were organ-
ized in groups, had received credit in 1970 from the institu-
tions participating in the Project, and had ready access to
information about the new maize recommendations; and
(b) farmers not on credit lists who were not organized, did
not receive credit from the participating institutions, and
had limited access to information about the new recom-
mendations. Two samples of farmers were interviewed in
the 1970 survey: (a) farmers on credit lists and (b) all
farmers in the area (a random sample of farmers from both
the above categories).
The survey of farmers on credit lists involved a stratified
random sample in which each stratum was made up of
farmers who had received credit from one of the following:
the National Agricultural Credit Bank, the National Ejidal
Credit Bank, the Agricultural Bank of the South, and the

Impulsora de Puebla. This stratified sample was used
because it was thought that there might be significant
differences among the characteristics of the groups of
farmers receiving credit from the four institutions.
The components of variance among and within the strata
were estimated using the maize yields of farmers on credit
lists in the 1970 season. (These yields had been estimated
using an indirect procedure described on page 76). These
components were used to determine the size of the sample
necessary to estimate averages in maize yields with a
90-percent probability. There were 213 farmers in the
A sample design in two stages, similar to that of 1967,
was used for the survey of all farmers in the area. The
components of variance among and within segments were
estimated using the 1970 maize yields of all farmers in the
area. The size of the sample of segments and number of
farmers per segment were calculated using these compo-
nents of variance, together with the quotient estimated in
1967 by dividing the average cost of identifying and taking
a census of a segment by the average cost of interviewing a
farmer. The number of segments was calculated at 25, but
was increased to 31 to provide greater precision in the
estimates. There were seven farmers per segment.
The sampling procedure used in 1967 assumed an unre-
stricted, random distribution of variables throughout the
Project area. This random selection of segments, however,
produced a pattern in which some parts of the area were
sampled with greater intensity than others. Between 1967
and 1970, geographical trends in the distribution of varia-
bility were discerned in the area. These trends showed the
assumptions of the 1967 sample to be faulty; the sample
left large areas unrepresented, whereas other areas were
sampled quite intensively.
Because of the above findings, plus the high costs
involved in taking the census of segments, the following
arbitrary scheme was developed for the first-stage sampling
in 1970.
Twenty-one of the segments identified in 1967 were
used in the 1970 survey. Ten new segments were added,
located at random in parts of the area not adequately
covered in 1967. Information about the Project area col-
lected between 1967 and 1970 was used in delineating the
10 zones in which the new segments were located. The new
segments were located in the field and the individual parcels
A listing was made of all farmers who worked at least
one parcel in the 31 segments. The sample of farmers was
selected at random from these lists and interviews were
made during the summer of 1971 by students from the
National School of Agriculture at Chapingo, who had
received 10 days of specialized training.
The completed questionnaires from the sample of farm-
ers on credit lists and the sample of all farmers in the area
were checked, and the information transferred to coded
sheets. Data processing was done at the Statistics and
Computation Center of the Graduate College, Chapingo.


Information relative to the infrastructure that had been
collected through interviews with farmers was supplemented
with data obtained directly from the agricultural institu-
tions in special studies done in 1968 and 1973.
The 1968 study focused on the level of services provided
by the credit banks, crop insurance company, and the
marketing agency. In addition, the fertilizer distribution
network was identified by compiling a list of all persons
who sold fertilizer in the villages of the area.
The objectives of the 1973 study were: (a) to obtain a
detailed understanding of the operating procedures and
policies of the three official credit banks, the Impulsora de
Puebla, and the crop insurance company; (b) to tabulate
the amount of services provided by these institutions during
the period 1965-1973; (c) to observe changes that had
occurred in the operating procedures of these institutions in
recent years; and (d) to determine to what extent such
changes had been influenced by the Puebla Project.
Changes in the operating procedures of the institutions,
and the amounts of services which they provided to farm-
ers, were studied by gathering information directly from
the institutions. Interviews were held with decision-makers,
both in the office and field, and additional information was
obtained from their files.
The Project farmers were interviewed in 1973 to obtain
a better understanding of why many farmers were not fol-
lowing the Project recommendations. Among the specific
issues covered in this study were: (a) the effect of farmer
organizations on the adoption of.the new technology, (b)
the level of understanding by the farmers of the maize
recommendations, and (c) the farmers' impressions of the
efficiency of the services provided by the agricultural
The farmers interviewed in 1973 were located in 10
communities distributed throughout the five Project zones.
These 10 communities were selected because they appeared
to represent the area adequately in terms of the attention
received from the Project technical assistance agents, and
from the agricultural service institutions in Puebla. The first
part of the study consisted of informal visits by the inves-
tigator to the communities over a period of 6 months to
establish rapport with the villagers. In the second phase of
the study a structured questionnaire was used to interview
samples of: (a) farmers in organized groups, and (b) unorgan-
ized farmers. The first sample included 69 farmers, 35 of
whom were group representatives and 34 were group
members. The sample of unorganized farmers consisted of
29 heads of families in five communities where organized
groups had functioned during the previous 3 or 4 years.


A major goal of the Puebla Project was to increase
production per unit area of maize; thus, an accurate and
continuing measurement of yields was crucial. Maize yields
could be measured by selecting a random sample of fields in
the area each year, with subsequent harvesting and measur-
ing of grain yields. This method required locating the fields,
locating the farmers (who usually lived some distance away
in a neighboring village), obtaining permission to harvest the
necessary sample area, harvesting in the presence of the
farmer, and returning the grain to the farmer. All of the
sample fields had to be harvested within a brief period-
from maturity of the earliest plantings to the beginning of
harvest by the farmers. These considerations prompted the
search for a simpler method for estimating maize yields.

Development of an Indirect Method
for Estimating Yields

In 1968, an estimation of maize yields was designed and
conducted to provide data for developing a simple, indirect,
reasonably precise method for estimating yields. In this
process, the length of ear filled with grain, diameter of the
unshucked ear at the base, and weight of grain with 12
percent moisture were determined for each ear in the maize
fields included in a random sample. A prediction equation
was developed by regressing grain yield-per-ear on the
diameter and length of the ear. This prediction equation
was used to prepare a table in which ear lengths were listed
as row headings, ear diameters as column headings, with
grain yields composing the body of the table. A rapid and
reasonably accurate estimate of yield can be made with this
table, using measurements of lengths and diameters of all
ears in a sample area of a maize field.
This table has been used since 1969 for making annual
estimates of maize yields in the Project area. A detailed
description of the statistical procedures used in developing
the regression model is given in an unpublished paper
(Heliodoro Diaz C., Delbert T. Myren, and Richard E.
Lund, "Estimating Corn Yields in the Puebla Area with a
Regression Model Based on Ear Length and Diameter").

Estimation of Yields of All Farmers in the Area

An annual estimation of maize yields was made for two
categories of farmers: (a), all farmers in the area and (b)
farmers on credit lists. In 1971 and 1972, in addition to
estimating yields, information was collected from the farm-
ers on the use of technology in the parcels included in the
A three-stage sample was used for estimating average
yields of all farmers in the area. In the first stage, the
segments selected for the 1967 and 1970 surveys were used.
In the second stage, a random selection of parcels was made
within the segments. In the third stage, five locations of 10
lineal meters each were chosen and distributed as shown in
Fig. 8.1.

The components of variance among segments, among
parcels, and within parcels were calculated annually, using
the maize yields estimated in the evaluation of yields. There
were 25 sample segments in 1968, 36 segments in 1969 and
1970, and 31 segments in 1971 and 1972. The number of
locations within a parcel remained constant from year to
year. The number of parcels per segment varied as a func-
tion of the magnitudes of the variances calculated for the
previous season. The selection of parcels was made using a
random systematic technique, with probability propor-
tional to size; that is, a 4 ha parcel had four times as much
probability of appearing in the sample as a 1 ha parcel.
After the parcels had been selected, the field was sam-
pled as follows. First, a coin was flipped to determine direc-
tion of entry into the field. Then the number of rows was
counted from left to right. Next, using a table of random
numbers, the starting row was selected. In this row, 10 out
of the first 20 meters were harvested, as shown in Fig. 8.1.
Then the sampling was moved five rows to the right, pro-
gressively, until a total of 50 meters had been harvested. If
this procedure carried the sampling to the outer edge of the
field, as shown in Fig. 8.1, the procedure was begun again
on the opposite edge and the counting of rows was resumed
toward the right.

100 m

Fig. 8.1. The diagram shows the way in which the field was
sampled for a total of 50 meters of harvested rows. From
each 20 meters of row, two sections of 5 meters each were
selected as shown.

In 1968, the yield estimate in each parcel was made by
harvesting all the ears of maize within the 50 lineal meters.
From 1969 onward, the yield estimate in each parcel was
made using the indirect procedure described above.

Estimation of Yields of Farmers on Credit Lists

The average maize yield of the 103 farmers who used the
Project recommendation in 1968 was estimated from yield
measurements made on each farm. On farms where a farmer
used the recommendations in two or more parcels, one
parcel was selected at random for sampling.
In 1969 and 1970, a random sample in three stages was
used for estimating the average maize yields. In the first
stage, the credit groups were the population, and a sample
of these groups was selected. In the second stage, a sample of
parcels within groups was drawn from among all the parcels
in selected groups for which the farmers had received credit
for using the new technology. In the third stage, sites
within the parcels were selected according to the scheme
illustrated in Fig. 8.1.
In 1971 and 1972, the first step in estimating average
maize yields was to divide the Project area into the five
work zones described in Chapter 5. Lists were prepared, by
zones, of all farmers on credit lists; farmers were randomly
selected from the five lists. Among those farmers in the
sample who received credit for using the recommendations
on only one parcel, this parcel was chosen for sampling.
Among farmers who received credit for two or more par-
cels, one parcel was selected at random. Sites within parcels

An indirect method was de-
veloped in 1968 for estimat-
ing the average maize yield in
the project area just prior to
harvest. Measurements of the
lengths, diameters and
weights of all ears from a
large number of plots were
used to calculate a regression
equation. Using this rela-
tionship, yields were estimated
in subsequent years by meas- .
during ear lengths and diame-
ters in a selected area of a b-
random sample of fields. ,i ,o

were selected as shown in Fig. 8.1. The number of farmers
per zone in the sample was determined from variances
calculated from estimates of yield made during the previous
year for each zone.
After the sites within parcels were selected, the estima-
tion of yield was made using the indirect procedure de-
scribed on page 76.

Comments on the Evaluation Program

Some deficiencies in the operation of the evaluation pro-
gram are now apparent. As mentioned, a primary concern
of evaluation was to quantify the Project's progress, and
primary emphasis was placed on the socioeconomic surveys,
the annual estimates of yield, and the use of this informa-
tion for evaluating change. As a result, much less impor-
tance was given to identifying obstacles limiting farmer use
of the new technology and in studying means to overcome
these barriers. Thus, the Project was sometimes slow in
modifying its operational strategies, particularly with
respect to farmer organizations and the service institutions.
This deficiency could be overcome by arranging for an
evaluation staff to receive assistance from highly trained
consultants with a broad understanding of evaluation. It
may also be necessary to increase the resources allocated to
evaluation and to provide the staff with additional training.
Another evaluation deficiency related to the rate at
which the collected and processed data was fed back to the
rest of the Project staff. Data on yield and the use of tech-
nology collected each year at harvest time were generally

made available within 2 months after the surveys were
completed. Information from the 1970 socioeconomic
survey, on the other hand, was not made available to the
staff until early 1973. This was due to a series of problems
related to coding, programming, and data processing that
could probably have been avoided with additional technical
Two methodological problems arose in evaluation; one of
these stemmed from use of the variance of maize yields as
the basis for estimating the sample size used in the surveys.
Since the objective of the surveys was to measure many
characteristics of the farmer, his family, and farm, it would
have been more appropriate to use the variance of a more
inclusive variable, such as total family income, in estimating
sample size.
A second methodological problem arose in selecting the
segments for the 1970 survey of all farmers in the area. The
arbitrary scheme that was followed was thought to offer a
better representation of the Project area. This scheme,
however, led to difficult theoretical problems in hypothesis
testing, inasmuch as there appeared to be no appropriate
way to estimate the variances of the statistics of the 1970
survey. Thus, two alternatives were available for analyzing
the changes that occurred in the area from 1967 through

1970: (a) to consider only the 21 common segments of
the two surveys. In this case, variance estimation would be
easy to compute and hypothesis testing would be straight
forward. However, the aggregated area of the 21 segments,
rather than the total Project area, would become the
population to which direct inferences could be made.
Inferences about the total Project area would have to be
guided by past experience and general knowledge; (b) to
consider all the information collected from the 25 segments
in 1967 and the 31 segments in 1970 in making inferences
about the total Project area. In this case, it would not be
appropriate to test hypotheses about population parame-
ters. Again, past experience and general knowledge would
play a role in comparing the statistics.
It was decided to adopt the second alternative for this
report. Thus, no variances or confidence intervals are
presented in the chapters where data from the 1967 and
1970 surveys are compared.

As mentioned, methods used for selecting the segments
used in the estimation of yields of all farmers in the area
differed from 1968 to 1969-1970, and to 1971-1972. This
should be taken into account in studying the yield data in
Table 9.8.




As described in Chapter 3, new recommendations on the
production of maize were available at the beginning of
1968, and these recommendations were modified for
subsequent years, particularly for certain parts of the area
and for late dates of planting. Promotion of the use of these.
recommendations was begun in 1968 with 103 farmers,
and was extended throughout Zones I, II, III, and IV in
1969, and to the remainder of the area in 1970.
In general surveys in 1967 and 1970, the Project evalua-
tion team obtained information useful for estimating the
extent to which the recommendations of the Puebla Project
have been used by the farmers in the area. Similar informa-
tion was obtained for the 5 years from 1968 to 1972 when
the evaluation team estimated maize yields just prior to
harvest and obtained interviews with the farmers whose
fields were sampled. This chapter presents some of these
data: (a) to show the degree to which farmers have used
the recommendations, and (b) to compare average maize
yields during 1967 to 1972. A final section discusses the
influence of certain factors on farmer adoption.


Quantitative discussion of the level of adoption of the
new maize recommendations is complicated by several
factors: (a) there was an initial tendency for farmers to
only partially adopt an individual production practice. For
example, instead of changing from a traditional level of
30,000 plants/ha to a recommended level of 50,000, the
farmer often changed to some intermediate level; (b) farm-
ers often tended to accept some recommended practices
more readily than others; that is, they might increase their
rate of nitrogen fertilization before changing the way they
apply it; and (c) farmers often tended to use the new tech-
nology initially on only a portion of their land.
Thus, the evaluation of the level of adoption became a
matter of determining the percentages of farmers who were
using the. various recommended practices in different
degrees at different times. Such evaluation in the Puebla
area was hampered by the diversity in recommendations
that had evolved over the years. By 1972, specific maize
recommendations were available for 16 producing condi-
tions. Recommended rates of nitrogen varied from 60 to
130 kg/ha for rainfed maize, rates of phosphorus from 0 to
60 kg P20 /ha, and plant densities from 30,000-50,000

plants/ha. The information collected for measuring adop-
tion rate, however, was taken from a sample of farmers
selected at random from the Puebia area as a whole, or
from the five geographical zones where the technical as-
sistance agents were assigned. Thus, information was not
available for individually evaluating the level of adoption of
the specific maize recommendations for the 16 producing
Lacking the above information, it was decided to estab-
lish arbitrary ranges in values of the several recommended
practices, corresponding to high, intermediate, and low
levels of adoption and apply them to the entire project
area. These ranges are shown in Table 9.1. The lower limits
for high levels of adoption of nitrogen, phosphorus, and
plant density--80 kg/ha, 30 kg/ha, and 40,000 plants/ha,
respectively-are the lowest rates of these inputs that were
being recommended in the area in 1973; except in the case
of nitrogen, for one producing system with a very low
production potential and a second system with plantings
made immediately following alfalfa; and, in the case of
phosphorus, for the two producing systems in Zone V for
which no phosphorus is recommended). (The limits be-
tween high and intermediate levels of adoption were estab-
lished as a function of the recommendations for unlimited
capital (Chapter 3). Thus, many of- the farmers in
intermediate category can be considered high adopters in
terms of Project recommendations for limited capital.) The
upper limits for the low levels of adoption correspond
approximately to rates used by farmers who were making
most intensive use of fertilizers in 1967. The phosphorus
ranges for Zone V are different from the rest of the area
because phosphorus has not been recommended for that
region since 1970.

TABLE 9.1. Ranges in rates of nitrogen, phosphorus and
plants per hectare corresponding to high, intermediate and
low levels of adoption of the three practices.
Phosphorus (kg/ha of P205) Plant density
Level of Nitrogen For Zones For Zone (thousands
adoption (kg/ha) I,II,III,IV V* per ha)

Low 0-50 0-20 > 30 0-30
Inter. 51-80 21-30 11-30 30-40
High >80 > 30 0-10 > 40

* The ranges in rates of phosphorus corresponding to the three levels of
adoption are reversed for Zone V, since phosphorus is not recom-
mended for maize in that region.

Most of the available information on the levels of use of
nitrogen, phosphorus, and plant density refers to a random
sample of parcels for the Project area. In general, therefore,
the analysis made here refers to the percentage of parcels
with a certain level of adoption of the recommended prac-
tices. If farmers were to use the new technology uniformly
on all their parcels, then the percentage of parcels with a
given level of adoption should be similar to the percentage
of farmers with the same level of adoption. However, in the
Project, where the farmers had an average of slightly over
three parcels and tended to adopt the new technology
initially on only a part of their land, it was expected that
percentages calculated in terms of parcels would be lower
than percentages calculated in terms of farmers for a given
level of adoption. This assumption was substantiated by
estimating (from the survey data for 1967 and 1970) the
percentages of farmers with a high level of adoption of
nitrogen and phosphorus, and comparing them with per-
centages of parcels with high levels of use of the two prac-
Analysis here is in terms of the average amounts of
nitrogen, phosphorus, and plants per hectare and to the
percentages of parcels on which these practices were used at
high, intermediate, and low levels. The available informa-
tion with respect to time of applying fertilizers, weed
control, and insect control was not sufficient for drawing
conclusions about changes in farmer use of these practices.

All Farmers in the Area

The 1967 survey involved a random sample of all the
farmers in the Project area. The 1970 survey, as well as the
yield evaluations for the years from 1968 to 1972, involved
a sample of all farmers in the area and another sample of
farmers on credit lists. (These farmers were organized into
groups and were aided by the technical assistance agents in
arranging for credit and in using the new technology
properly.) The data from the sample of all farmers provide
the relevant information on the level of adoption of the
recommendations for the Project area and are presented first.
The use of the recommendations by the farmers on credit
lists is discussed later.
The average rates of nitrogen and phosphorus and
average plant densities for maize plantings in the Puebla
area for the period 1967-1972 are shown in Table 9.2.

TABLE 9.2. The average amounts of nitrogen and phos-
phorus in kilograms per hectare and the average number of
plants in thousands per hectare used in maize plantings in
the Project area from 1967 to 1972.
Practice 1967 1968 1969 1970 1971 1972 %

Nitrogen 34 -- 53 83 78 129
Phosphorus 14 -- -- 19 30 27 93
Plant density 31 31 33 33 34 10

The difference between the values in 1972 and the first year with
information, expressed as a percentage of the value for the first year.

From 1967 to 1972, the average increases were: nitrogen,
129 percent; phosphorus, 93 percent; and plants/ha, 10
percent. The increase in nitrogen use is remarkable and
reflects the farmers' general awareness of the need to apply
large amounts of this fertilizer. The slightly smaller change
in the average application of phosphorus is due in part to
the recommendation that farmers use no phosphorus for
maize in Zone V.

The small change in average plant densities is believed to
be due to one or more of the following reasons: (a) farm-
ers are often uncertain at planting time whether they will
obtain all the fertilizer they need; they use a rate of seeding
lower than that recommended with the idea that the maize
will produce better at the lower plant density, should they
not obtain sufficient fertilizer; (b) farmers are concerned
about drought and believe their maize will do better, in case
of drought, if the plant density is low; and (c) the farmers'
major concern is in increasing their production of large ears,
and they feel this can best be achieved with plant densities
below the recommended levels. All of these reasons have a
certain validity and provide an excellent example of how
difficult it is to convince low-income farmers in rainfed
areas to radically change their technology, and how dif-
ficult it is to develop and deliver recommendations that are
adequate for the extremely variable production and
economic conditions of the farmers.

The levels of adoption of the recommended rates of
nitrogen, phosphorus and plant density can also be exam-
ined in terms of changes in the percentages of parcels with
high, intermediate, and low levels of use of these practices.
Table 9.3 shows the information needed for this analysis for
the period 1967-1972. During this time, the percentages of
parcels with a high level of adoption of the three practices
increased; whereas, the percentages of parcels with a low
level of adoption decreased, and the percentages of parcels
in the intermediate category remained constant.
In 1972, the percentages of parcels with a high level of
adoption of each of the three practices were about equal,
varying from 44.8 percent for nitrogen to 39.4 percent for
plant density.
From 1967 to 1972, the increase in the percentage of
parcels with a high level of adoption of the nitrogen recom-
mendation (37.4 percent) was much greater than the in-
creases for the phosphorus (20.6 percent) and plant density
(25.4 percent) recommendations. This finding again sug-
gests that the farmers in the area have accepted the nitrogen
recommendation more readily than the phosphorus and
plant density recommendations. The 25.4 percent change in
the percentage of parcels with a high level of adoption of
the plant density recommendation indicates a greater
acceptance of this recommendation than was suggested by
the increase of only 10 percent in the average plant density
for all plantings in the area.
Since there is usually a positive interaction among the
production factors in their effects on maize yields, the
maximum increase from a package of production practices
is obtained when all factors are used at the recommended

TABLE 9.3. Percentages of parcels in the Project area with high, intermediate and low levels of adoption of the nitrogen,
phosphorus and plant density recommendations.

Level of Year
Practice adoption 1967 1968 1969 1970 1971 1972 Change*

High 7.4 33.1 33.6 44.8 + 37.4
Nitrogen Intermediate 11.0 -14.0 15.0 14.0 + 3.0
Low 81.6 -- 52.9 51.4 41.2 40.4

High 23.7 -- 38.4 44.4 44.3 + 20.6
Phosphorus Intermediate 7.7 -- 9.2 6.1 9.1 + 1.4
Low 68.6 52.4 49.5 46.6 22.0

High 14.0 15.8 24.9 23.8 39.4 + 25.4
Plant density Intermediate 35.2 34.0 30.8 29.0 33.9 1.3
Low 50.8 50.2 44.3 47.2 26.7 24.1

* Change is the difference in the values for 1972 and the first year in which information was available.

TABLE 9.4. Percentages of parcels in the Project area with
all combinations of high, intermediate and low levels of
adoption of the nitrogen and phosphorus recommenda-
tions. 1967 1970 1971 1972
Levels of adoption (N=337)* (N=713) (N=214) (N=221)

High for both practices 4.8 20.7 19.6 29.9
High for one;
intermediate for the
other 7.7 10.5 12.2 6.8
High for one;
low for the other 13.9 19.4 26.6 22.6
for both practices 0.3 1.0 0.0 0.9
Intermediate for one;
low for the other 10.4 10.8 8.9 14.5-
Low for both 62.9 37.6 32.7 25.3
N is the number of parcels in the sample.
levels. It is enlightening, therefore, to examine the degree to
which the nitrogen, phosphorus, and plant density recom-
mendations have been adopted simultaneously.
Information is available for 1967, 1970, 1971, and 1972
on the percentages of parcels with all combinations of the
three levels of adoption of the nitrogen and phosphorus
recommendations. Table 9.4 shows the percentage of par-
cels with high levels of adoption of both nitrogen and phos-
phorus increased from 4.8 percent in 1967 to 29.9 percent
in 1972. The percentage of parcels with a low level of adop-
tion of both practices decreased during the same period by
a larger amount, from 62.9 percent down to 25.3 percent.

Information on the percentages of parcels with the three
levels of adoption of the nitrogen, phosphorus, and plant
density recommendations is available for 1971 and 1972
only. Table 9.5 shows a high level of adoption of the three
practices on 10.4 percent of the parcels in 1972. There was
a high level of adoption of at least two of the practices on
41.2 percent of the parcels in the same year.
The data in Tables 9.3 through 9.5 show a relatively high
percentage of the parcels with a high level of adoption of

individual recommended practices, a lower percentage of
parcels with a high level of adoption of the nitrogen and
phosphorus recommendations, and a still lower percentage
of the parcels with a high level of adoption of the three
practices. This indicates that most farmers in the Puebla
area are presently not realizing the full potential of the
increased production that comes from using all the produc-
tion practices at the recommended levels. Clearly, the job
of adjusting and delivering adequate technology, as well as
that of inducing farmers to use the recommended tech-
nology, is very difficult, and is far from being accomplished
in the Puebla area.

TABLE 9.5. Percentages of parcels in the Project area with
different combinations of high, intermediate and low levels
of adoption of the nitrogen, phosphorus and plant density

1971 1972
Levels of adoption (N = 214)* (N = 221)

High for the three practices
High for two; intermediate for one
High for two; low for one
High for one; intermediate for two
High for one; intermediate for one;
low for one
Intermediate for the three practices
Intermediate for two; low for one
High for one; low for two
Intermediate for one; low for two
Low for the three





* N is the number of parcels in the sample.

Farmers on Credit Lists

Farmers on credit lists receive credit for purchasing the
inputs, mainly fertilizers, required for the recommendations
of the Puebla Project. The technical assistance agents
provide information about amounts of fertilizers to apply,
when and how to apply them, the seeding rate, and other
recommended practices. As a group, farmers on credit lists
would be expected to use the recommendations of the
Puebla Project most efficiently.
Table 9.6 shows the numbers of farmers on credit lists
and the corresponding areas of maize for which credit was
received during the years 1968 to 1973. As noted previ-
ously, both the number of farmers and the area for which
credit was received increased rapidly in 1969 and 1970,
with a slower rate recorded for the following 3 years. In
1973, 16.6 percent of all the farmers in the Puebla area
received credit for the production of maize according to
Project recommendations, representing 25.8 percent of the
total harvested area of this cereal. (Thus, the percentage of
the area for which credit was received is about 50 percent

TABLE 9.6. The number of farmers on credit lists and the
areas of maize for which credit was received in 1968-1973.

Year No. of %/ of Area % of
farmers total* ha total**

1968 103 0.2 76 0.1
1969 2561 5.9 5838 7.3
1970 4833 11.1 12601 15.8
1971 5240 12.1 14438 18.0
1972 6202 14.3 17533 21.9
1973 7194 16.6 20604 25.8

Based on a total of 43,300 farmers.
** Based on a total of 80,000 ha of maize.

TABLE 9.7. Percentages of parcels of farmers on credit lists
nitrogen, phosphorus and plant density recommendations.

greater than the percentage of farmers receiving credit.
This, however, does not necessarily imply that the larger
farmers have greater access to credit. It is known, for exam-
ple, that some farmers include the needs of other members
of their family in their request for credit.)
Table 9.7 shows the percentages of parcels of farmers on
credit lists with high, intermediate, and low levels of adop-
tion of the nitrogen, phosphorus, and plant density recom-
mendations. The percentages for nitrogen and phosphorus
in 1971 and 1972, and the percentages for plant density for
all years refer specifically to parcels for which credit was
received. The percentages for nitrogen and phosphorus in
1970 refer to all parcels of the farmers on credit lists. The
differences between 1970 and 1971 in the adoption of the
nitrogen and phosphorus recommendations probably over-
state the change in adoption occurring that year; they also
indicate that some of the farmers on credit lists apply the
recommended practices on only a part of their parcels.
In 1971 and 1972, about 75 to 80 percent of the parcels
of farmers on credit lists for which credit was received
showed a high level of adoption of the nitrogen and phos-
phorus recommendations; judgment of the adequacy of this
level of adoption of the fertilizer recommendations by
farmers on credit lists should take into account that five
technical assistance agents were assisting 5,240 farmers in
1971 and 6,202 in 1972.
In 1968, 82 percent of parcels of farmers on credit lists
showed a high level of adoption of the plant density recom-
mendation; this figure declined to 36.1 percent in 1970 and
has remained fairly constant. The high level of use of the
plant density recommendation in 1968 was due to the close
supervision (particularly at planting time) of the 103 farm-
ers on credit lists by one full-time and one part-time tech-
nical assistance agent. The percentage drop in 1969 and
again in 1970 probably reflects the rapid increase in the
number of farmers on credit lists and the resulting decline
in the assistance that could be given to each farmer. The
low percentage of parcels in 1970 through 1972 with a high
level of adoption of the plant density recommendation

with high, intermediate and low levels of adoption of the

Level of Ye__
Practice adoption 1968 1969 1970 1971 1972 Change

High 51.1 72.9 75.8 + 24.7
Nitrogen Intermediate 25.5 19.9 17.7 7.8
Low -23.4 7.2 6.5 16.9

High 64.4 82.3 76.6 + 12.2
Phosphorus Intermediate -7.6 5.0 9.7 2.1
Low 28.0 12.7 13.7 14.3

High 82.0 55.7 36.1 29.8 37.1 -44.9
Plant density Intermediate 18.0 34.4 28.7 28.7 34.7 + 16.7
Low 0.0 9.9 29.0 41.5 28.2 + 28.2

Change is the difference in the values for 1972 and the first year in which information was available.

indicates the previously discussed (page 80) reluctance of the
farmers in the Puebla area to use high plant populations.
Perhaps this reluctance is a reflection of the fact that
information received by the farmer is imperfect, or that the
farmers adjust the recommendations in terms of their
perception of application to their local conditions.


Information on average maize yields in the Puebla area
was available for the years 1967 through 1972. Estimations
of yield were made directly in the field from 1968 to 1972,
using the method described in Chapter 8. Maize yields also
were estimated for 1967 and 1970 from information col-
lected from farmers in the surveys, taking into account all
the parcels on which the interviewed farmers grew maize.
The average maize yields for all farmers in the area and
for farmers on credit lists are shown in Table 9.8. Using
1968 as a base, the changes in average yields for the fol-
lowing years were calculated and are shown in the table as
percentages. The year 1968 was used as a base, rather than
1967, because it was felt that the Puebla Project could not
have influenced the general average for 1967 and because
all estimations of yields were made in the same way begin-
ning in 1968.
The average maize yields for farmers on credits lists
(Table 9.8) varied from 3,985 kg/ha in 1968 to 2,679 kg/ha
in 1971. The high average in 1968 can be attributed to very
favorable rainfall conditions, and to the fact that the yields
of only 103 carefully selected farmers (who received close
supervision by the technical assistance agents) entered into
the calculation. The average yields of farmers on credit lists
varied little from 1969 through 1972.
The average maize yields for all farmers in the area have
varied from 1,330 kg/ha in 1967 to 2,499 kg/ha in 1972.
Comparing only the average yields for the first and the last
years, it is seen that the average yield increased by 88 per-
cent. This, however, overestimates the real increase in maize

TABLE 9.8. Average maize yields* for all farmers in the
Puebla area and for farmers on credit lists.

All farmers
in the area

Year kg/ha



Farmers on
credit lists

% change
to 1968

- 8.3
- 9.9



% change
to 1968


Grain with 14%moisture. The value for 1967 was cal-
culated from information provided by farmers in the
survey; values for the other years were calculated from
field measurements made just prior to harvest.
** The average yield for 1970, calculated from the survey
data, was 1864 kg/ha.

yields in the area, because rainfall conditions were much
more favorable in 1972 than in 1967.
In a rainfed area like Puebla, average maize yields in a
given year are determined largely by the climatic conditions
that prevail and the production technology that is used. To
estimate the effect of the new technology on average yields
in the Puebla area, it is necessary to adjust the average
yields in Table 9.8 by eliminating the effect of climate.
Two methods have been used to estimate the percent
increase in average maize yields of all farmers in the area
due to use of the production practices recommended by the

Average maize yields for the
project area were adjusted for
the effects of climate using
yield data from fertilizer rate
experiments conducted each
year in the area. The yields of
plots receiving a uniform
treatment in the several ex-
periments conducted each
year were averaged, and the
variation in these yearly aver-
ages was assumed to be due
to climatic differences.

TABLE 9.9. An estimation of the increase in average maize yields in the Puebla area due to the use of the recommended
production practices, in which the effect of climate is calculated from experimental data.

(a) (b) (c) (d) (e)
Average Estimation of the Estimated yields Differences in % increase in
yield for effect of climate assuming no yield in kg/ha average yields
Year all farmer (%change change due to attributable to attributable to
kg/ha compared to new technology** new technology new technology
1968)* kg/ha (a-c) d x
c 100)
1967 1330
1968 2140 2140 0
1969 1832 -18 1755 + 77 4.4
1970 1962 -15 1819 +143 7.9
1971 1927 -21 1691 +236 14.0
1972 2499 6 2011 +488 24.2

* These percentages were calculated from the average yields obtained in the field experiments with the treatment consisting
of 50 kg/ha of nitrogen, 25 kg/ha of P205 and 30 thousand plants per hectare.
" 2140+(the value in column b) (2140), where 2140 is the average yield in 1968.

Puebla Project. The first of these involves the use of yield
data from the fertilizer rate experiments conducted each
year in the Project area. Data were available from 8 to 12
experiments in each of the years from 1968 to 1972.
Average yields were calculated for the plots in the several
experiments receiving 50 kg/ha N, 25 kg/ha P205, and
30,000 plants/ha. (This treatment was used because it
produced average yields similar to those for all farmers in
the area. Since an interaction can be expected between
production level and climatic effects, it was desirable that
the average levels of production of the selected treatment
and all the farmers be similar.) The changes in these average
yields with respect to 1968 were calculated for the years
1969-1972. Shown in column b, Table 9.9, these changes,
expressed as percentages, are estimations of the effect of
climate. These percentages were multiplied by the average
yield in 1968 to obtain the differences in yield due to
climatic effects. Then, the differences in yield due to cli-
mate were added to the average yield in 1968 to obtain the
average annual yields unaffected by the new technology, as
shown in column c, Table 9.9.
The differences between the average yields for all farm-
ers and the estimated yields assuming no effect of the new
technology were considered to be the effects attributable to
the use of the new technology. These differences are shown
as percentages in column e, Table 9.9.
As shown in Table 9.9, estimated increases in average
yields of all farmers in the area varied from 4.4 percent in
1969 to 24.2 percent in 1972. This method for adjusting
average yields for the effect of climate has obvious defi-
ciencies. The number of experiments that provided the data
for this calculation was too small to sample the area ad-
equately. Also, these experiments were not distributed over
the Project area so as to give proper weight to the 16 pro-
ducing systems.

The second method for adjusting average yields for the
effect of climate used the information obtained in the
objective yield measurements of samples of farmers on
credit lists. It was assumed that those farmers on credit lists
with yields in the upper third of the sample had used the
recommended technology quite adequately, and that this
level of use of the technology had been reasonably constant
over the years. It was further assumed that (for a given
year, using 1968 as a base) the change, in the average yield
for the upper third of farmers on credit lists was a measure
of the relative favorableness of the climate for that year.
This method was used to estimate the effects of climate
for the years 1969-1972 with the results shown in column
b, Table 9.10. Data for farmers on credit lists in Zone V
were not included in this calculation, because the Puebla
Project did not begin to promote the use of new technology
in that region until 1970. The estimated combined effects
of climate and use of the new technology (Table 9.8) are
reproduced as percentages in column c, Table 9.10. The
percentage increases in yield with respect to 1968 at-
tributable to the use of the recommended practices, shown
in column d of Table 9.10, were calculated by subtracting
the effect of climate, (column b) from the combined effects
of climate and technology (column c). The estimated
increases in yield due to use of the new technology (column
f) were calculated by multiplying the percentage increases
in column d by the average maize yield in 1968. The aver-
age yields without the new technology were estimated
(column g) by subtracting the increases in column f from
the average yields for the area in column e. The increases in
yield due to the new technology, expressed as a percentage
of the average yields without the technology, are shown in
column h.
According to this second method of adjusting average
yields for the effect of climate, estimated increases in aver-

TABLE 9.10. An estimation of the increase in average maize yields in the Puebla area due to the use of the recommended
production practices, in which the effect of climate is calculated from the maize yields of farmers on credit lists.
(a) (b) (c) (d) (h)
Average yield Estimation of Estimation of the Estimation of (e) (f) (g) ,%increase in average
of the upper the effect of effect of climate the effect of Average Estimated increase Estimated average yields attributable
1/3 of farmers climate plus technology technology (c-b) yields for in yield due to yield without the to technology
on credit lists (%change (%change (%change area technology** technology (e-f) f
Year kg/ha compared to 1968) compared to 1968)* compared to 1968) kg/ha kg/ha kg/ha x 100)
1968 4965 base base -2140 -2140
1969 4090 -17.6 -14.4 + 3.2 1832 68 1764 3.9
1970 4085 -17.7 8.3 9.4 1962 201 1761 11.4
1971 4043 -18.6 9.9 8.7 1927 186 1741 10.7
1972 4087 -17.7 + 16.8 + 34.5 2499 738 1761 41.9
FromTable 9.8
** Percentage in Colurin (d) multiplied by the average yield in 1968.

age yields of all farmers in the area varied from 3.9 percent
in 1969 to 41.9 percent in 1972. This method also has
obvious limitations. The average use of the new technology
by the upper third of farmers on credit lists may have been
higher in 1968, than in other years. Or, stated more general-
ly, there is no empirical basis for assuming that the use of
technology by the upper third of farmers on credit lists
was reasonably constant. It is also possible that the upper
third of farmers on credit lists does not provide a repre-
sentative sample of the producing conditions in the Puebla
The two methods for estimating the increases in maize
yields attributable to the use of the new technology give
similar percentages for 1969, 1970, and 1971, but differ
markedly for 1972. It is probably reasonable to assume that
the true percentage increase in average yields due to the
new technology is somewhere near the average of the values
obtained with the two methods. This calculation would
suggest that average maize yields in the Puebla area prob-
ably increased through the use of improved technology by
about 30 percent from 1967 to 1972.
The above efforts to adjust average maize yields for the
effect of climate indicate the need for a project to develop
plans from the outset for collecting the data required for
such an adjustment. Experience in the Puebla Project sug-
gests that the necessary data can be generated by making
simple plantings, consisting of three plots managed at low,
medium, and high production levels, at sites distributed
throughout the Project area. The number of sites required
would be determined as a function of the variability among
sites, and these would be located adequately to sample the
different producing systems in the Project area. The same
general sites (but not the exact site) and plot treatments
would be used each year. If sufficient information were not
available the first year to accurately establish the limits of
the several producing systems, the number of sites should
be increases initially by perhaps 100 percent to assure that

each system was adequately sampled. It should be possible
at a later date (once the limits of the producing systems
were defined) to reduce the number of sites to those neces-
sary, based on the variability among sites and the level of
precision desired. Annual differences in the average yields
for each of the treatments should provide a reliable esti-
mate of the effects of climate.


Some information on the adoption of the maize recom-
mendations, such as the number of farmers on credit lists
(Table 9.6), suggests a rapid rate during 1969 and 1970 and
a somewhat slower rate for the next 3 years. The informa-
tion on the increases in average yields attributable to the
new maize technology (Tables 9.9 and 9.10), on the other
hand, indicates an accelerated rate of adoption in 1972.
Based on the available information, it seems reasonable to
conclude that there has been a fairly continual rate of
increase in the use of the Project recommendations since
This increase in the use of the new maize technology has
produced an increase in average maize yields that has been
estimated to be around 30 percent over the 4-year period
1969-1972, or about 7.5 percent per year. There seem to be
no valid yardsticks for judging whether this is a reasonable
rate of progress for a rainfed area with a moderate level of
agronomic risk. It is evident, however, that many farmers at
the end of 1972 were not using the recommendations (41
percent of parcels with a low level of adoption of the
nitrogen recommendation, Table 9.3), and others were only
using them partially (75 percent of parcels with a low level
of adoption of one or more of the three main practices,
Table 9.5). Thus, it seems appropriate to ask why the rate
of adoption has not been faster and to examine some of the
reasons farmers have continued to use their traditional

Availability of Information

In promoting the use of the Project's new recommend-
ations, the technical assistance agents also told the farmers
how to apply them; what they would cost to use; the
expected increases in production and net income from their
use in good, average, and poor years; and the importance of
using each practice at the recommended level. It was as-
sumed that the farmers (particularly those who provide the
leadership for the community) would require full knowl-
edge of the new technology to make accurate appraisals.
In examining the extent to which information on the
Project recommendations has been disseminated through-
out the Puebla area, it is important to distinguish be-
tween: (a) a simple understanding of what the recom-
mendations are and (b) full knowledge of how to use them
and of the expected returns in terms of increased produc-
tion and net income.
Relevant data on farmers' knowledge of the recom-
mendations was collected in 1973, in a study of farmers not
on credit lists in five communities where groups of farmers
organized by the Puebla Project had functioned for 3 or 4
years. Of the 29 farmers interviewed, 26 (90 percent) had
heard of the maize recommendations of the Puebla Project.
Only 15 (52 percent) of the 29 farmers, however, were
convinced that the use of the maize recommendations
would result in higher yields.
These data suggest that by 1973 most of the farmers in
the Puebla area had heard of the new maize recommenda-
tions. A much smaller percentage, however, perhaps around
50 percent, had received information sufficient to persuade
them that the new technology would increase yields. The
low level of use of one or more of the three main practices
on 75 percent of the parcels in 1972 (Table 9.5) suggests
that perhaps 25 percent or less of the farmers understood
the more complex aspects of the new technology, such as
the importance of using all of the recommended practices
at the recommended levels. Clearly, the Project recom-
mendations have not been completely understood by the
farmers, thus preventing their full realization of the poten-
tial benefits of the new technology.

Adequacy of the New Technology

Another interpretation can be made regarding the 48
percent of the farmers in the 1973 survey who were not
convinced of\ the usefulness of Project recommendations:
that rather than an indication of the lack of adequate
information, it could be that the new technology is, in fact,
not superior to the traditional practices. Certainly, lack of
adequate technology has been a notable weakness of many
programs seeking to improve agricultural production in
rainfed areas.
The adequacy of the maize recommendations of the
Puebla Project was examined in some detail in Chapter 3.
Table 3.11 shows that producing systems 1.1.1, 2.1.1, and
3 (which account for 53 percent of the cultivated area in

maize) had estimated average maize yields using the tradi-
tional technology of: 2.05 ton/ha (1.1.1); 2.15 ton/ha
(21.1); and 2.56 ton/ha (3). The estimated average yields
using the Project recommendations for unlimited capital
were 3.80 ton/ha (1.1.1); 3.87 ton/ha (2.1.1); and 3.64
ton/ha (3). For the entire Project area, the estimated aver-
age yields were 2.05 ton/ha using traditional practices and
3.13 ton/ha using the Project recommendations for unlim-
ited capital.
The estimated net incomes from using the traditional
and Puebla Project technologies, expressed in kg/ha of
maize, are shown in Table 3.13. The estimated net incomes
using the two Project recommendations are larger than the
estimated net incomes using the traditional practices in
each of the 16 producing systems. For the entire area, the
estimated net incomes using the Project recommendations
were 51 percent greater for limited capital and 95 percent
greater for unlimited capital, as compared to the estimated
net incomes using the traditional practices.
Another indication of the adequacy of the Project
recommendations is that most farmers, after they have used
the new technology, apparently continue to use it in the
following years. This is a reasonable conclusion to draw
from the findings that both the level of use of the recom-
mendations and the average maize yield in the area have
increased at a fairly constant rate during the period
1969-1972. Had a significant proportion of the farmers in
the area realized lower net incomes because of inadequacy
of the new technology, it seems reasonable that the use of
the recommendations and the average yields would have
leveled off or declined by 1972.

Risk in Using the New Technology

In a rainfed area such as Puebla it can be argued that the
average expected increases in yield and net income from the
use of the new technology are not as important to the
farmer as is the probability that net income using the
Project recommendations may be less than with the tradi-
tional practices in some years. It seems reasonable that the
major concern of many low-income farmers is to assure an
adequate food supply in very unfavorable years; i.e., their
first concern is in maximizing the probability of covering
family needs in poor years, rather than maximizing average
yields and net income.
To the extent that this sort of decision-making occurs, it
can be expected that small farmers will accept or reject the
new technology in terms of their perception of how it will
influence their net income in an unfavorable year. Although
the Project did not directly measure farmers' perceptions of
the risk involved in adopting the new technology, some
appreciation of the importance of such risk can be obtained
from information collected during the 6-year period
As described in Chapter 3, net incomes from the use of
several production strategies were calculated from the
results obtained in 125 fertilizer rate-plant density experi-

ments conducted during 1967-1972. These net incomes
were used to estimate the risks farmers take in using the
several technologies. Risk was defined arbitrarily as the
standardized probability of obtaining an increase in net
income from the use of a given technology equal or inferior
in value to: (a) 0.5 ton/ha of maize grain or (b) 0 ton/ha
of maize grain.
As shown in Table 3.14 for traditional technology, risk
defined as the probability of a net income of 0.5 ton/ha or
less, was nearly four times as great in producing system
1.1.1.; three times as great in system 2.1.1; and 32 percent
greater in the entire area-as compared with that using
recommendations for unlimited capital. When defined as
the probability of a net income of 0 ton/ha or less, risks
using the traditional practices were nearly three tir es as
great in producing system 1.1.1; twice as great in system
2.1.1; and about 4 per cent less for the entire area-as
compared with that of using Project recommendations for
unlimited capital.
Comparison of net incomes and risks using Project
recommendations for unlimited capital and those for tradi-
tional practices suggests several observations about the
relative risks involved: (a) for average and favorable years
there is a high probability of an attractive income from
using either technology; the expected net income is nearly
twice as large with the Project recommendations as with
traditional practices; (b) for less favorable years, the value
of the net income will be equal to or less than 0.5 ton/ha
of maize in many instances; the probability of these low
incomes is much higher with the traditional than with the
recommended technology; (c) for the least favorable years,
net incomes less than zero can be expected: the probability
of net losses is similar for the two technologies; and (d) net
incomes using the traditional technology are sometimes (12
percent of the experiments during 1967-72) larger than the
Project recommendations. Based on the available informa-
tion it appears that farmers, as a whole, would assume less
risk by using the Project recommendations than by using
the traditional technology. There are instances, nonetheless,
where farmers will lower their net incomes by switching
from the traditional to the new technology. Hopefully, the
frequency of these latter situations can be gradually re-
duced as agronomic knowledge of the area is improved.

It seems quite likely that the above evaluation of the
significance of risk differs from farmers' perception of the
risk involved in using the new technology. The results
farmers have obtained from using the Project recommenda-
tions have been less favorable than those reported in the
experiments, which can be attributed to failure to use the
recommendations fully, as discussed earlier. The farmers
have probably encountered a higher percentage of cases
(higher than the 12 percent cited previously) where the
Project recommendations have been less profitable than the
traditional practices. Also, the experience of the farmers
extends over a much longer span of years than the period
covered in this study, and almost certainly includes years
less favorable than any of those of the 1967-1972 period. It

seems reasonable to conclude that a certain percentage of
the farmers in the area, perhaps one-fourth, now feel that a
change from the traditional to the new technology would
mean a reduction in net income from their maize in the less
favorable years.

Availability of Credit

As can be calculated from the information in Table 3.10,
the average cost involved in using the Project recommenda-
tions for limited capital is 19 percent greater than for the
traditional technology; the average cost of the recom-
mendations for unlimited capital is 82 percent greater than
for the traditional technology. According to the data col-
lected in the area, only about 15 percent of the farmers
have sufficient personal funds to purchase the inputs re-
quired by the Project recommendations. Thus, in deciding
whether to use the new technology, the greater fertilizer
cost and the necessity for credit to cover this expense is a
major consideration for most farmers.
The discussion of the credit institutions in Chapter 7
suggests that the supply of credit available through the
official banks for maize production has been equal to or
greater than the demand in recent years. A study of 29
farmers not on credit lists from five communities in the
Puebla area in 1973 indicated that there were two main
reasons why more farmers did not request credit from one
of the official banks: (a) 15 of the farmers (52 percent)
feared that they would not be able to pay back the loan-
they were particularly concerned about the possibility of an
unfavorable year, and about their lack of understanding of
the whole process of obtaining credit; and (b) seven of the
farmers (24 percent) were repelled by the number of re-
quisites they had to meet to qualify for a loan-theywere
particularly bothered by having to pay a premium for crop
insurance that they felt provided no real protection.
From the standpoint of the farmer, it appears that the
lack of satisfactory access to available credit is limiting his
adoption of the Project recommendations. Supporting
evidence for this conclusion was presented in Chapter 7-the
demand for credit from the Impulsora de Puebla, which can
be arranged for very simply and without crop insurance, has
been far greater than the supply. Hopefully, this obstacle'
can be overcome, both through favorable change in the
credit institutions and by increasing farmers' knowledge of
the operation of the banks and the adequacy of the recom-

Farmer organizations

Recognizing the farmers' credit needs for the purchase
of fertilizers, as well as the difficulties individual farmers
have in securing loans from the banks, the Project team
began in 1969 to assist small farmers to organize ana wuLr
together as organized groups. This activity of the Project
team has been viewed as an essential part of the operational

strategy of an agricultural program where the ratio of small
farmers to technical assistance agents is very large. This
emphasis on farmer organization has increased the number
of farmers in the area who have been able to obtain credit
for maize production. The questions can be asked, how-
ever: Are the requirements for becoming members of a
group too difficult? Would adoption of the Project recom-
mendations increase if such requirements were less restric-
Information relating to these questions was collected in
1973 in interviews with 69 farmers belonging to 35 groups
distributed throughout the area. Approximately 65 percent
of the farmers indicated that the only requirement for
becoming a member of their group was that the candidate
be an honest, responsible person who fulfilled his obliga-
tions. About 20 percent of those interviewed indicated that
candidates had to deposit some piece of property with the
representative of the group to guarantee that they would
pay back the loan at the end of the year. This latter require-
ment at first appeared restrictive. However, investigation
revealed that it was necessary only that the candidate, if
considered to be honest and responsible, sign a contract
with the group in which he agreed to repay the loan at the
end of the year.
Clearly, those farmers who are judged by their neighbors
to be dishonest and irresponsible are unlikely to become
members of the farmer organizations. Apart from these,
however, there was no clear evidence that the requirements
for membership in the groups constitute a factor limiting
farmer adoption of the maize recommendations.

Other Factors

Characteristics of the farmer, his family, and his land
(such as level of education, size of the family, number of
members of the family that work, family capital resources,
farm size, quality of the land, etc.) probably influence the
farmer's decision to adopt or not adopt the Project recom-
mendations. With the information available it was not
possible to determine the importance of the first four of
these factors on adoption.
The 1970 survey data, however, were used to study the
influence of farm size on farmer use of the maize recom-
mendations. The amount of nitrogen used by the 50
percent of the farmers with the largest farms was compared
with the amount used by the 50 percent of the farmers
with the smallest farms. On the average, the farmers with
the larger holdings used 41 percent more nitrogen per
hectare than the farmers with the smaller holdings.
Quality of land almost certainly influences the adoption
of new technology. Farmers recognize differences in the
potential of lands to produce and are more likely to use
expensive technology on land with high-yielding potential.
It was not possible to study this factor in Puebla as infor-
mation on land quality of the sampled individual holdings
was not available.
Other factors which often influence the adoption of new
technology are the relationships between input costs and
product prices, availability of inputs, and the network of

roads in the project area.
Obviously, the more favorable the relationship between
maize prices and fertilizer costs in Puebla, the more likely
that farmers will adopt Project recommendations. It seems
unlikely, however, that prices have been an important
factor restricting farmer use of the new technology. The
relative prices of maize and fertilizers have remained fairly
constant during the 1968-1972 period. Approximately 4 kg
of maize remained equal in value to 1 kg N; and 3 kg maize
to 1 kg P2 05. As noted in Chapter 3, net income from the
use of fertilizers is quite favorable in most of the producing
systems in the area, with this price relationship.
Fertilizers were sold in some 46 towns and villages in the
area during the early years of the Project. Since 1971, most
fertilizers have been distributed through the official banks
and franchised dealers in six major towns. In most in-
stances, however, farmers have been able to purchase the
quantities of fertilizers they need. On occasion, however,
they have had to wait several weeks for fertilizer deliveries
and have not always been able to buy the materials they
prefer. Difficulties in purchasing fertilizers have probably
had some influence on the adoption of the maize recom-
mendations, because: (a) farmers who were lukewarm
about the use of fertilizers, have decided to use less ferti-
lizer in the face of inconveniences in procurement, (b)
those who have been unable to purchase fertilizers prior to
planting have, at times, reduced their rate of seeding and
thus have obtained lower returns from the fertilizers; and
(c) those who have received and applied fertilizers later
than recommended have sometimes been disappointed with
the results.
Difficulties in transporting fertilizers and produce have
probably not influenced the rate of adoption of the Project
recommendations. As mentioned in Chapter 1, there is an
adequate system of roads in the Puebla area.

The organizing of farmers into groups has helped to in-
crease the number of small farmers that have been able to
use project recommendations.

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