155
Necesidades de investigación futuras
Francisco J. Morales
Virólogo, CIAT
Gracias al trabajo de todos los profesionales y científicos dedicados al 
estudio y control de la mosca blanca Bemisia tabaci y de los geminivirus que 
esta especie transmite, hemos logrado acumular un conocimiento extenso 
sobre estos problemas. La red Mesoamericana y del Caribe, coordinada por 
el Dr. Luko Hilje de CATIE, ha logrado aunar esfuerzos y diseminar 
información en toda la región. El Proyecto Suizo (COSUDE) de 
PROFRIJOL, hizo igualmente posible el desarrollo de variedades resistentes 
al mosaico dorado-amarillo en la América Latina. Finalmente, El Proyecto 
Global de Mosca Blanca, financiado por la Agencia de Desarrollo Danesa 
(DANIDA), y coordinado por la Dra. Pamela K. Anderson desde el CIAT, 
ha realizado un trabajo exhaustivo de investigación y compilación sobre 
Bemisia tabaci y los geminivirus que atacan cultivos básicos e industriales 
en el mundo, considerando el volumen de información generado. Ahora 
necesitamos proyectos que nos permitan hacer un análisis intensivo de toda 
la información recogida y, más importante, necesitamos aplicar los 
conocimientos adquiridos a nivel de campo.
Una de las áreas más afectadas por las crisis económicas que han venido 
afectando a la América Latina, ha sido el mejoramiento genético. Son pocos 
los cultivos afectados por geminivirus en el mundo, donde se pueda mostrar 
tanto progreso en la generación de variedades resistentes, como en el caso 
del frijol. Existen aún algunos programas de mejoramiento de frijol que 
buscan mejorar esta leguminosa por su resistencia a virus transmitidos por 
mosca blanca, como son los de la Escuela Agrícola Panamericana 
(Zamorano) en cabeza del Dr. Juan Carlos Rosas; el de la Universidad de 
Puerto Rico, a cargo del Dr. James Beaver; el del Norte de México, llevado 
por el Ing. M.Sc. Rafael Salinas; y el del Proyecto Frijol, en el CIAT, 
liderado por los Drs. Shree Singh y Steve Beebe. Es necesario continuar el 
apoyo y reforzar estos trabajos de mejoramiento convencional, que tantos 
éxitos han tenido.
El uso de técnicas moleculares para el desarrollo de variedades de frijol 
resistentes a geminivirus, es promisorio, y felizmente cuenta con recursos
156
económicos adecuados. Es necesario poner esta metodología al servicio de 
los programas nacionales y mejoradores de frijol en general.
Los trabajos pioneros de la Universidad de Wisconsin, coordenados por el 
Dr. Douglas Maxwell, han permitido vislumbrar la posibilidad de 
transformar plantas de frijol buscando su resistencia a geminivirus. Este ha 
sido un trabajo difícil, que ya encuentra eco en otros países de la América 
Latina, donde científicos como el Dr. Josias Faria, ha transformado frijol 
para el control del mosaico dorado del frijol. Igualmente se reconoce la labor 
que viene desarrollando el Dr. Robert Gilbertson, Universidad de California, 
Davis, para la comprensión de los mecanismos moleculares que gobiernan la 
interacción entre geminivirus y sus hospederos.
La epidemiología de los virus transmitidos por mosca blanca en frijol, ha 
sido el tema de los trabajos realizados por la Dra. Pamela K. Anderson, en 
Nicaragua y luego en el CIAT, Colombia. Actualmente se encuentra 
trabajando en el mejoramiento de un modelo matemático que nos permitirá 
tomar decisiones de control según los parámetros existentes en cada región 
afectada por estas plagas. Estos estudios se han beneficiado de la 
información que constantemente generan investigadores que trabajan a nivel 
de campo y/o laboratorio, como la Dra. Jane Polston y el Dr. Emest Hiebert 
de la Universidad de Florida, o la Dra. Judith K. Brown, en la Universidad 
de Arizona. Estos científicos han estado constantemente involucrados con 
los problemas de la producción de alimentos causados por geminivirus en la 
región mesoamericana.
Se inició también en el CIAT, un estudio sobre la distribución espacial de las 
enfermedades causadas por geminivirus transmitidos por mosca blanca, 
utilizando sistemas de información geográfica (SIG). Estos estudios nos 
permitirán analizar mejor los factores que intervienen en el desarrollo de 
epidemias de geminivirus en diversos cultivos y regiones geográficas.
Por último, necesitamos desarrollar un sistema de monitoreo y de alerta 
sobre la posibilidad de ataques severos de geminivirus y mosca blanca, 
basados en los estudios y análisis epidemiológicos y espacio-temporales.
Nuestra mayor esperanza está basada en un gran número de investigadores 
latinoamericanos, los cuales se han formado en áreas relacionadas al 
conocimiento de los geminivirus y de los biotipos de mosca blanca que 
afectan los cultivos en la América Latina. Solo esperamos que se encuentre
157
un balance entre los investigadores de campo y de laboratorio, para llevar a 
feliz término la misión que les ha sido encomendada: mejorar la 
productividad de los cultivos afectados por geminivirus, en especial los 
cultivos alimenticios básicos, como el frijol.
159
English summary
Francisco J. Morales 
Virologist, CIAT
Current situation of common bean production and whitefly 
transmitted begomoviruses affecting Phaseolus vulgaris in 
Latin America
Common bean research and production in Latin America
The common bean (Phaseolus vulgaris L.) remains one of the most 
important food crops in Latin America, where this legume provides rural and 
urban populations with most of the essential protein and carbohydrates. 
However, common bean production in Latin America has drastically 
declined in the past decade, as a response to the globalization of the 
economy. The new economic policies have resulted in decreasing financial 
support from national and international governments to agricultural research 
in Latin America. The limited resources available are being used to conduct 
research on non-traditional export crops or natural resource management, 
often, with a minimum research capacity that needs to be supplemented with 
special project funds. As a result, most of the main common bean producing 
countries in Latin America are currently importing this basic food to meet 
the internal demand. In Latin America, Argentina has become the main 
supplier of common beans, producing approximately 300,000 tons a year on 
average. Other common bean producing countries are Chile, Ecuador and 
Bolivia, but the total area devoted to the production of common beans for 
export only amount to approximately half a million hectares, whereas the 
total land abandoned to common bean production in Latin America in the 
last decade, exceeds a million and a half hectares.
Another negative aspect of common bean production in Latin America, is a 
notable decrease in the capacity to generate new varieties. Without strong 
and functional common bean breeding programs, Latin America will not be 
able to meet current or future research challenges posed by the continuous 
emergence of new biotic and abiotic production problems. Although the use 
of molecular techniques, such as molecular markers, is expected to 
compensate for the downsizing of common bean breeding programs, Latin
160
America still has a long way to go before these techniques are effectively 
used for common bean improvement purposes. Likewise, the successful 
model of common bean programs conformed by teams of scientists from 
different disciplines, has also disappeared from most national programs and 
even international institutions devoted to agricultural research in Latin 
America.
Begomoviruses infecting common bean and their control
In the meantime, the emergence of new geminiviruses and whitefly biotypes, 
continues to threaten the viability of common bean production in the 
lowlands and mid-altitude valleys of Latin America. Whereas Bean golden 
mosaic virus (BGMV) and Bean golden yellow mosaic virus (BGYMV) 
remain the most important geminiviruses of common bean, other 
geminiviruses transmitted by Bemisia tabaci have shown their capacity of 
adaptation and destmctive potential in common bean. A clear example is the 
adaptation of Squash leaf curl virus (SLCV) to common beans in 
northwestern Mexico, and the resulting yield losses caused by the common 
bean variant of SLCV, known as Bean calico mosaic virus (BCaMV). In 
recent surveys of northwestern Mexico, conducted by the author, strains of 
SLCV can still be recovered from mosaic-affected common bean plants. 
This virus (SLCV) probably disseminated from northwestern Mexico and/or 
southwestern United States, and is now found affecting different 
cucurbitaceous crops in Central America, posing a threat to bean production 
in this important common bean-producing region.
Fortunately, private industry is currently financing some of the research 
activities abandoned by national programs, and we still have a few active 
and experienced bean breeders, who continue to generate excelent common 
bean varieties possessing resistance to different geminiviruses. This is the 
case of Ing. Agr. M.Sc. Rafael Salinas, common bean breeder located in Los 
Mochis, Sinaloa (N.W. Mexico), who has developed common bean cultivars 
highly resistant to BCaMV. Curiously, he has been able to use some of the 
sources of resistance identified for other begomoviruses of common bean, 
particularly red kidney genotypes of Andean origin.
Bean golden yellow mosaic virus (BGYMV) has its domains from southern 
Mexico, affecting all the Central American and Caribbean countries down to 
Colombia. Dr. Julio Bird first used this name to refer to the striking 
yellowing symptoms induced by a geminivirus (later selected as the type
161
strain of BGMV) in Phaseolus lunatus and P. vulgaris, in Puerto Rico. 
Although this species has probably evolved since its original molecular 
characterization in the last decade, the most noticeable change in some 
isolates of BGYMV, has been their loss of reactivity with a monoclonal 
antibody selected seven years ago to distinguish BGYMV from other 
begomoviruses infecting common bean in Latin America (FJ. Morales, 
unpublished results).
The control of BGYMV in southern Mexico, Central America and the 
Caribbean region has been possible thanks to the international and national 
breeding for resistance project on BGYMV initiated in the mid-1970s in 
Guatemala. The BGYMV-resistant lines generated by this project have made 
an exceptional impact throughout this region, and even in South American 
countries, such as Argentina, where these lines have been equally resistant to 
the distinct Bean golden mosaic virus species. Although the output of this 
project has been significantly reduced due to the current financial situation, 
other projects, such as the Bean/Cowpea (CRSP) project has been 
successfully complementing these efforts, particularly in Puerto Rico (Dr. 
James Beaver, University of Puerto Rico) and in Honduras (Dr. Juan Carlos 
Rosas, Escuela Agrícola Panamericana).
Another potential threat to common bean production in the Caribbean 
region, is the presence of an Old World geminivirus introduced into the 
Dominican Republic from Israel: Tomato yellow leaf curl virus (TYLCV). 
This virus has already been isolated from diseased common beans in Spain, 
and the common bean production regions in Mesoamerica are usually close 
to the tomato growing areas affected by TYLCV. Fortunately, there seems to 
be a larger number of potential sources of TYLCV resistance in Phaseolus 
vulgaris than in tomato.
In South America, with the exception of Colombia, where BGYMV has 
been detected, the predominant species is Bean golden mosaic virus 
(BGMV). This virus was first reported in Brazil by Dr. Alvaro Santos Costa 
in the early 1960s, and later shown to be different to the “bean golden 
mosaic” viruses described in Mesoamerica (BGYMV and BCaMV). BGMV 
arrived in Northwestern Argentina in the early 1980s, and, during the last 
decade, to the department of Santa Cruz, Bolivia, following a horizontal 
dissemination pattern in the savanna regions of South America located 
between 10° and 30° latitude south. Another gemini virus that has caused 
significant yield losses in this region, particularly in northwestern
162
Argentina, has been Bean dwarf mosaic virus (BDMV). This vims, related 
to Abutilón mosaic virus (AbMV), was first described in Brazil, together 
with BGMV, but its incidence in Brazil was usually low. In the late 1970s, 
BDMV (or a closely related vims) caused severe yield losses in common 
bean fields located in northwestern Argentina. BDMV has been found in 
Colombia and in Nicaragua, infecting beans at a relatively low incidence. 
Other geminivimses have been isolated from common bean in northwestern 
Argentina, including Tomato yellow vein streak, described later on from the 
State of Sao Paulo, Brazil.
The incidence of BGMV in the common bean production region of Santa 
Cmz, Bolivia, is relatively low at this time, mainly because some of their 
bean fields are at altitudes above 1,000 meters. At this altitude, the 
predominant whitefly species associated with common bean is Trialeurodes 
sp. and not the vector Bemisia tabaci. Nevertheless, the increasingly 
frequent phenomenon of “El Niño”, which creates unusually dry conditions 
in some Andean highland regions, has made possible the increasing 
multiplication of B. tabaci populations above 1,000 m.
The development of BGMV-resistant common bean cultivars has been rather 
slower in South America than in Mesoamerica, due to different reasons. In 
Brazil, institutions such as CNPAF (National Center for Research on 
Common Bean and Rice) and LAPAR (Agronomic Institute of Parana), have 
developed a number of BGMV-resistant lines. Additionally, many BGYMV- 
breeding lines developed in Central America, have been taken to Brazil, 
Argentina and Bolivia, where they have performed well under BGMV 
pressure, suggesting the existence of similar mechanisms of resistance to 
BGYMV and BGMV. These breeding lines and BGYMV-resistant cultivars 
have also performed very well under BDMV pressure.
Geminivirus taxonomy
The taxonomy of geminiviruses has changed rather dynamically in the last 
decade, mainly, driven by advances in molecular techniques that facilitate 
the amplification of geminiviral DNA and its rapid sequencing. These 
techniques are now being used to characterize geminiviruses transmitted by 
Bemisia tabaci in Latin American countries, such as Mexico, Guatemala, 
Honduras, Costa Rica, Panama, Colombia, Venezuela, Brazil and Argentina.
163
Currently, the majority of geminiviruses affecting cultivated and wild plants 
in Latin America, are transmitted by Bemisia tabaci. All of these viruses are 
considered New World viruses, and are native to the American tropics and 
subtropics, where B. tabaci can thrive. The only exception is Tomato yellow 
leaf curl virus, introduced into the Americas from Israel. Considering that B. 
tabaci was introduced into the Americas, probably from Asia, it is 
understandable that whitefly-transmitted geminiviruses are also found 
outside the Americas, namely in Asia and Africa. However, the type strain 
of the newly recognized genus Begomovirus, which groups all of the 
whitefly-transmitted geminiviruses known world-wide, is the South 
American geminivirus Bean golden mosaic virus (BGMV), from which the 
genus derives its name. BGMV is considered a unique species, distinct from 
the geminivirus that induces similar symptoms on common bean throughout 
Mesoamerica, recently renamed Bean golden yellow mosaic virus 
(BGYMV), as it was first named by Dr. Julio Bird in Puerto Rico. The third 
distinct geminivirus species found in the Americas, causing “golden 
mosaic”-like symptoms on beans, is Bean calico mosaic virus (BCaMV), a 
more distant virus closely related to Squash leaf curl virus (SLCV).
Dr. Maria del Rosario Rojas has contributed a chapter on the molecular 
characteristics of begomoviruses, where she describes four filogenetic 
groups of bean geminiviruses, including the Mesoamerican BGYMV 
isolates, the South American BGMV, Bean dwarf mosaic virus (BDMV) 
from South and Central America, and BCaMV from N.W. Mexico.
Drs. Douglas Maxwell, Stephen Hanson, Josias Faria and Robert Gilbertson 
discuss the molecular techniques currently used by most researchers to 
detect and identify begomoviruses. Monoclonal antibodies (developed at the 
University of Florida under the direction of Drs. Ernest Hiebert and Dan 
Purcifull), specific and broad spectrum probes, and PCR, are the main 
diagnostic techniques used today by most laboratories. These techniques 
have been used to identify the main wild reservoirs of common bean 
begomoviruses, but, so far, only few wild hosts have been confirmed as 
such. The University of Wisconsin has also been a leading institution in the 
implementation of antiviral strategies and common bean transformation 
methods, based on the study of the genomic function of begomoviruses. 
Among the antiviral strategies mentioned are: capsid protein- mediated plant 
resistance, creation of non-functional Rep proteins, trans-dominant lethal 
rep gene constructs, and antisense strategies.
164
The whitefly vector
The discussion on the whitefly vector is led by Dr. Pamela K. Anderson, 
who points out the need to consider Bemisia tobad not only as an insect pest 
but as a virus vector as well. Whitefly population management is necessary 
to reduce the intensive use of pesticides, and to prolong the useful life of the 
begomovims-resistant common bean cultivars developed by national and 
international institutions in Latin America.
The identification of B. tabad has also been facilitated by the development 
of taxonomic keys for both immature and adult specimens, thanks to the 
work of Dr. Rafael Caballero in Honduras (EAP). Dr. Anderson also 
discusses the current controversy among taxonomists, regarding the 
identification of the biotype B of B. tabad as a distinct species named 
Bemisia argentifolii (Perring & Bellows). Whereas this possibility is not 
discarded, most whitefly taxonomists feel that further research is needed to 
settle this controversy. In the mean time, the name B. tabad biotype B is 
retained in this publication. Nevertheless, the higher reproductive rate and 
broader host range attributed to biotype B of B. tabad, requires further 
investigation from the epidemiological point of view.
The biology and ecology of B. tabad is also described with special emphasis 
on the role of common bean as a reproductive host. Data collected in 
Mexico, Central America and Colombia, show that common bean is not a 
preferred reproductive host of B. tabad, although this species can reproduce 
abundantly on common bean in the absence of more suitable reproductive 
hosts, as observed in Brazil.
Regarding the epidemiology of begomoviruses transmitted by B. tabad, Dr. 
Anderson refers the reader to the chapter contributed in the first edition of 
this publication, produced in 1994, for complementary information. Here, 
she points out the need to identify the reproductive hosts of B. tabad in each 
common bean production region affected by begomoviruses, because these 
hosts are not necessarily the same in all countries. She lists over 30 different 
plant species identified in Latin America as hosts to B. tabad. Finally, Dr. 
Anderson lists the main categories of pesticides used in Latin America to 
control B. tabad, and emphasizes the need to implement IPM practices to 
reduce the use of synthetic pesticides in common bean production regions. 
One of the most promising methods of control discussed, is the use of
165
biological control agents, namely entomopathogens and predators of B. 
tabaci.
The identification of B. tabaci biotypes A and B, is the subject of the chapter 
contributed by Dr. Lee Calvert. He has implemented the RAPD-PCR 
method described by Dr. Paul De Barro in Australia, to differentiate the 
introduced biotype B from the local B. tabaci biotype. Dr. Calvert shows 
that the primers recommended by De Barro, can be used to differentiate 
biotypes A and B of B. tabaci in the Americas, but points out critical 
technical procedures and other considerations to be taken into account for 
the correct analysis of results. One of the main sources of error found, was 
the presence of other whitefly species on common bean in the Americas, 
particularly Trialeurodes vaporariorum, which can produce similar RAPD 
patterns with at least one of the primers used.
Integrated begomovirus and whitefly management
By far, the most successful approach to the control of geminiviruses 
affecting common bean, has been breeding for disease resistance. The 
implementation of this approach was possible thanks to the efforts of a 
number of donors, particularly US and Swiss donors, who funded CLAT and 
national programs in Mexico, Central America and the Caribbean region, to 
start a very successful breeding project in Guatemala. The first generation of 
BGYMV-resistant materials originated from a few black-seeded parental 
materials of Mesoamerican origin. One of these early lines, DOR-41, 
became a widely-adopted cultivar in Central América, the Caribbean, and 
even northwestern Argentina. Similar materials, were also adopted in 
southern Mexico. A second generation derived from the same sources of 
resistance to BGYMV, crossed to local landraces, yielded some excellent 
lines, such as ICTA-Ostúa, still under cultivation in Guatemala. The project, 
however, had to tackle a more difficult task: to breed for resistance to 
BGYMV in grain types other than black, mainly the highly prized red- 
seeded materials consumed in El Salvador, Nicaragua, Costa Rica and some 
Caribbean countries.
A breakthrough occurred when a non-black-seeded genotype, A 429, which 
had never been bred for resistance to BGYMV, showed unexpected high 
levels of resistance to this virus under field conditions. This line, originally 
selected for its superior plant architecture, was thoroughly evaluated at 
CIAT, together with its parental materials, to identify the source(s) of
166
resistance. Following their BGYMV screening under controlled conditions, 
two parental genotypes were identified as sources of BGYMV resistance in 
A 429: Porrillo Sintético (a black-seeded resistance source used for the first 
generation of DOR lines), and “Garrapato”, a rustic pinto variety of Mexican 
origin. The latter genotype was infected and had no pods due to the 
infection by BGYMV, but it did not show any yellowing symptoms. This 
rather questionable bean genotype, has become one of the main sources of 
BGYMV/BGMV resistance known to date.
Later on, a line selected for its resistance to BGYMV (derived from Porrillo 
Sintético), DOR 303, showed some interesting reactions to the virus under 
field conditions. Basically, it did not have noticeable yellowing symptoms, 
but a few plants were frequently affected by severe stunting. A similar 
examination of this line and its parental materials at CLAT, revealed the 
presence of red kidney types of Andean origin, which condition the type of 
resistance to BGYMV displayed by this line, in combination with P. 
Sintético.
A thorough search of bean genotypes available at the CLAT’s bean 
germplasm bank, led to the evaluation of selected accessions in different 
countries of South America, Central America, the Caribbean, and in Mexico. 
Some additional 15 bean genotypes were added to the sources of 
BGMV/BGYMV resistance previously identified. Genetic studies conducted 
under the direction of Dr. Shree P. Singh at CLAT, made possible the 
combination of various mechanisms of virus resistance identified in bean 
genotypes belonging to different races of Phaseolus vulgaris. Additionally, 
research conducted at the University of Puerto Rico, led to the identification 
of specific genes responsible for the responses observed in selected sources 
of resistance to the inoculation of bean begomoviruses. This research 
resulted in the development of molecular markers/techniques (RAPD, QTL, 
SCAR, microsatelites) to expedite the breeding for BGMV/BGYMV 
resistance work (marker assisted selection).
Chemical control is still widely practiced in Latin America to control the 
whitefly-transmitted geminiviruses that affect common bean. However, 
considering the relatively low economic resources of most bean farmers in 
this region, the types of pesticides applied to control B. tabaci, are usually 
the compounds more likely to be deactivated by this whitefly species, 
generating an increasing problem of development of pesticide-resistant 
whitefly populations. The new chemicals, such as imidacloprid, are very
167
effective but equally costly for Latin American bean farmers. Consequently, 
the use of non-synthetic pesticides and biocontrol agents is gaining terrain in 
Latin America, where botanicals, such as neem and tobacco extracts; 
predators, such as Eretmocerus and Encarsia spp.; and entomopathogens, 
such as Verticillium lecanii and Beauveria bassiana, are now being tested 
and applied more frequently. The use of some soaps to control B. tabaci, is 
another interesting method of control practiced by some small-scale farmers.
Cultural practices have not been widely adopted by bean farmers to control 
begomoviruses. The use of physical barriers, mainly anti-whitefly screens or 
screenhouses; live barriers or trap crops; sticky traps, planting density and 
planting dates, are seldom used in Latin America. Perhaps the most 
successful control practice implemented in certain countries of Latin 
America to control BGYMV, has been the enforcement of legal decrees 
banning the continuous cropping of whitefly-rearing hosts throughout the 
year. Such legal measures have been successfully implemented in the 
Dominican Republic, to break the B. tabaci-BGYMV cycle.
Future research needs
We hope that the breeding-for-resistance work that has been so successful in 
the past, will not be abandoned but, rather, expedited by marker assisted 
selection. We must make a concerted effort to maintain the existing 
whitefly/geminivirus research networks. There is the need to continue the 
basic research that is the foundation of the antiviral strategies, and to 
improve the available technology for common bean transformation. Finally, 
we need to advance our understanding of the ecology and epidemiology of 
whitefly-transmitted geminiviruses in different cropping systems. We 
sincerely hope that the information presented in this publication will help 
researchers in developing and industrialized countries to better define their 
research priorities.
169
Agradecimientos
Esta publicación no hubiera sido posible sin el concurso de todos los autores 
que contribuyeron su tiempo y conocimiento a su producción. La 
financiación de esta publicación ha sido posible gracias a la Agencias 
Danesa y Suiza de Cooperación para el Desarrollo (DANIDA y COSUDE, 
respectivamente). El Proyecto Suizo de PROFRIJOL ha sido la columna 
vertebral de esta publicación desde su primera edición. Agradecemos las 
gestiones del Dr. Rogelio Lepiz, ex-coordinador de esta red, para iniciar el 
proceso de recopilación de la información suministrada por los países 
miembros de PROFRIJOL. Se agradece igualmente al Dr. Cesar Cardona, 
Coordinador del Proyecto Frijol del CIAT, por su apoyo a esta publicación. 
Finalmente se reconoce la ayuda de los Srs. Julio Cesar Martinez, 
Guillermo Guzmán y de la Sra. Patricia Zamorano en la elaboración y diseño 
de este documento.
Acknowledgements
This publication has been possible thanks to the concourse of all the 
colleagues that contributed their time and knowledge. Also, the financial 
support of the Danish Agency for International Development (DANIDA) 
and the general support of the Swiss Agency for International Development 
(COSUDE) to the System-Wide IPM Project and the Mesoamerican bean 
research network, respectively, are gratefully acknowledged. Among the 
various collaborators in this undertaking, we would like to express our 
gratitude to Dr. Rogelio Lepiz, ex-coordinator of PROFRIJOL. Special 
thanks to Dr. Cesar Cardona, Bean Project Manager at CIAT, for his 
personal support to this publication. Finally, we are grateful to Mr. Julio 
Cesar Martinez, Guillermo Guzmán, and Mrs. Patricia Zamorano for their 
technical help in the production of this document.