Conservation of coconut (Cocos nucifera L.) biodiversity in Sri Lanka
Common names of the sweetpotato (lpomaea batatas) in the Americas
Insect pollination and isolation requirements in tomato collections (Lycopersicon esculentum Mill.)
Use of computer databases to manage an in vitro collection of potato germplasm
Genetic contamination by windborne pollen in germplasm-regeneration plots of smooth bromegrass
Genetic resources of patented seeds available to the public worldwide
L. Perera, R.R.A. Peries and W.M.U. Fernando
Coconut Research institute, Lunuwila, Sri Lanka
Summary
The progress in research toward the conservation of coconut biodiversity in Sri Lanka and the utilization of germplasm for the development of improved cultivars is reviewed. Both random and biased collections were made but preference was given to collecting populations that exhibit characteristics of economic importance. A total of 39 accessions has been collected to date and conserved in duplicate in ex situ genebanks. Consideration of long-term physiological adaptation of the palms led to development of in situ genebanks to produce planting material suitable for low-input sustainable agriculture systems. Procedures to quantify the between- and within-population variability among coconut populations using quantitative variation and molecular methods are also being investigated.
Introduction
Coconut is the most widely grown plantation crop in Sri Lanka, occupying about 416 000 ha of the total area of about 6.5 million ha. It is important in the daily diet of the average Sri Lanka, and has an estimated per capita consumption of 110 nuts/year. Around 80% of the estimated total production is consumed domestically. About 135 000 persons are directly or indirectly employed in the coconut sector which includes processing, trade and other related activities (CDA 1993).
The available coconut germplasm in Sri Lanka is categorized into three distinct varieties: typica, nana and aurantiaca (Liyanage 1958). Characteristics of different forms of these three varieties are shown in Figs. 1 and 2. The typica or tall type has been the most widely exploited. It is generally cross-pollinating, does not have a reliable method of vegetative propagation and has a long generation interval.
Systematic collecting and conservation of coconut germplasm were initiated in Sri Lanka around 1984 (Wickramaratne 1984). Previously there had been only a traditional varietal collection that was planted in 1940 and 1960. This collection was comprised of eight forms of the typica variety, three forms of the nana variety and two forms of the aurantiaca variety (Table 1) and the pool of elite palms at the Isolated Coconut Seed Garden (ISG) (Liyanage 1961ab).
Systematic genetic conservation was needed because of genetic erosion in coconut, which occurs at the rate of about 1% per annum of the area under coconut cultivation (Peries 1993). Exotic germplasm for breeding is not available because of the risk of introducing lethal diseases. The main reason for this high rate of genetic erosion is rapid industrialization and urbanization in the traditional coconut growing areas. Further, natural disasters such as cyclones and droughts have caused severe losses in valuable coconut genetic material amounting to about two million
coconut palms in the east coast area of Sri Lanka in 1978 and nearly 200 000 palms in the southern part of the country during 1988-89. The introduction of improved coconut varieties (Liyanage 1956, 1957) into the national replanting programme of the country has led to the gradual replacement of existing biodiversity over time.
- Fig. 1. Shapes of different forms of coconuts found in Sri Lanka: Forms of typica variety: A1-typica, A2-Kamandala, A3-bodiri, A4-navasi, A5-ran thembili, A6-gon thembili, A7-pora pol, A8-dikiri pol; forms of nana variety: B1-dwarf green, B2-dwarf yellow, B3-dwarf red; forms of aurantiaca variety; C1-thembili (King coconut), C2-navasi thembili (Liyanage 1958).
- Fig. 2. Longitudinal sections of nuts of coconut forms illustrated in Fig. 1. The legend is the same as that for Fig. 1 (Liyanage 1 958).
Table 1. Varieties and forms of coconut in Sri Lanka (according to Liyanage 1958)
Variety |
Breeding habit |
Forms |
Typica (tall) |
Out-breeding |
Typica |
Bodiri | ||
Ran thembili | ||
Gon thembili | ||
Nawasi | ||
Pora pot | ||
Kamandala | ||
Dikiri pol | ||
Nana (dwarf) (kundira) |
In-breeding |
Pumilla (dwarf green) |
Eburnea (dwarf yellow) | ||
Regia (dwarf red) | ||
Aurantiaca |
In-breeding |
Thembili (king coconut) |
Nawasi thembili |
Collecting: physiological adaptation and preferential selection
Although large populations of cross-pollinating species are likely to contain a large proportion of the total genetic variation, particularly for quantitative characters, it is unlikely that any single population will contain all of the genetic variation. The total genetic variation of a species is therefore likely to be fractionated over populations as the impact and direction of natural selection varies from one to another, due to environmental variation and genetic drift (Lawrence and Rajanaidu 1985).
It is reported that coconut existed in Sri Lanka as far back as 101-77 BC. However, growing coconut on a plantation scale commenced only after foreign invasion of the country during 1855. There is a great deal of circumstantial evidence to suggest that the coconut palm adapts itself to pests, disease and climatic factors associated with a particular location when grown over a period of time (Foale 1991). Although Sri Lanka is a small island, there is a great deal of climatic variation within the country which would generate a wide range of variation among the coconut populations, particularly the survival capabilities, depending on the diversity of the agro-ecology.
Evidence suggests that the coconut plantations in Sri Lanka, especially the large ones, have undergone preferential selection for various characters such as nut yield, shape, colour and size. In some instances, even the weight of husked nut and thickness of the kernel have been considered in the selection of mother palms over several generations. Therefore it can be reasonably assumed that populations in different environments and under different management are physiologically adapted to their respective agro-ecologies and thus have the capacity to represent diverse gene pools.
Under the circumstances, a part of the collecting programme was directed at representative samples from phenotypically different coconut populations. Historical information available in estates was reviewed and specific populations were identified for initial collecting rather than collecting at random from the entire island.
Exotic material
Prior to implementation of quarantine regulations to control the accidental introduction of coconut diseases, some coconut growers imported certain forms of coconut (Wickramaratne 1987; Peries 1991). Collecting activities are being focused on locating and purifying exotic material that has been built up through introduction and has become naturalized under different agro-ecological conditions in Sri Lanka.
Drought tolerance
Because of adverse weather and the resulting fluctuations in coconut yield, drought tolerance became a priority in coconut breeding in the mid-1980s. During 1988-89, the country, especially the southern part, experienced a prolonged drought that resulted in a loss of about two million coconut palms. In these areas, as well as in other areas such as Puttalam and Kurunegala situated in the periphery of the main coconut triangle where frequent droughts are common, certain populations and even individual palms maintained their productivity despite the severe water deficit. Such naturally adapted palms and/or populations were identified and representative samples collected as part of the germplasm conservation activity.
Other distinguished palms populations
The collecting programme also focused on individual palms and populations with phenotypes uncommon to the known diversity. The brown dwarf of the nana variety (Peries 1991) was one such interesting collection and spicata types (Wickramaratne 1987) were also included in this category.
Conservation: ex situ genebanks
Collected material has been conserved in ex situ genebanks representing two different environments (Table 2). To date these ex situ collections contain a total of 39 accessions comprising 13 traditionally classified varieties and forms, 21 tall accessions of the variety typica and four dwarf accessions including two exotic dwarf types.
Table 2. Current (after 1984) status of germplasm collection in Sri Lanka
Type |
Accession name |
Size of population/location_ |
Remarks_ | |
B/E |
PRS |
|||
Tall | ||||
Moorock |
84 |
82 |
ET | |
Palugaswewa |
60 |
85 |
ET | |
Pitiyakanda |
86 |
85 |
ET | |
Clovis |
85 |
85 |
NE | |
Namalwatta |
79 |
95 |
LEI/DT | |
St. Annes |
- |
85 |
ET | |
Margaret |
- |
85 |
NE | |
Kasagala |
- |
85 |
LEI/DT | |
Debarayaya |
- |
87 |
LEI/DT | |
Akurassa |
84 |
89 |
NE | |
Ambakelle special |
78 |
91 |
IV | |
Melsiripura |
- |
91 |
ET | |
Mangala Eliya |
- |
86 |
ET | |
Goyambokka |
- |
90 |
LEI/DT | |
Keenakelle |
- |
86 |
ET | |
Maliboda |
- |
87 |
ET | |
Horakelle |
- |
83 |
ET | |
Walahapitiya |
- |
85 |
ET | |
Wellawa |
84 |
79 |
DT | |
Ambakelle tall |
86 |
- |
IV | |
Goluwapokuna |
- |
90 |
||
West African tall |
3 |
- |
Exotic | |
Dwarf | ||||
Kundasale |
- |
88 |
ET | |
Braune (dwarf brown) |
- |
89 |
||
Cameroon red dwarf |
6 |
86 |
Exotic | |
Brazilian green dwarf |
5 |
20 |
Exotic | |
_B/E=Bandirippuwa estate in wet-intermediate zone, PRS=Poththukulama Research Station in dry-intermediate zone.
_ET=Ecotype, LEI=Low external-input system, DT=Putative drought tolerance, IV=lmproved variety, NE=Exotic, now naturalized.
In situ genebanks
In situ collecting was initiated according to the ecobank concept proposed by Peries (1993). The rationale of the in situ genebank is that the populations in a particular agro-ecology, subjected to mass selection over several generations by ordinary farmers and/or through natural selection, are distinguishably different from each other in survival capabilities associated with physiological adaptation, although they may be morphologically similar. Therefore the different populations that exist within the country contain beneficial genes that give the palms the much needed survival capability under a given farming system. Evidence suggests that 'improved' material introduced into these agro-ecologies may not be able to adapt rapidly enough to thrive under the same suboptimal agro-ecological systems.
Under this genebank programme, populations which are putatively adapted physiologically to have a high-yield potential are identified and conserved in situ after very poor yielders are eliminated from the population (Zobel and Talbert 1984). It is proposed that areas having similar agro-ecological conditions be replanted with seedlings from these in situ genebanks.
Evaluation
Without a detailed understanding of the genetic structure of the species and its populations, it is impossible to adopt a sampling strategy that captures a desired level of genetic variation. Because the genetic structure of coconut has not been studied adequately, the present collecting strategy has been drawn up to collect one nut from 50 to 100 palms per population for collection purposes (Carpio 1990) regardless of the fraction of total genetic variation among populations and among families within population components. Therefore, as a first step, a systematic trial was designed to exploit the between-population variation and the between- and within-family variation to develop suitable collecting strategies for our conservation programme in addition to evaluation and characterization methods.
The systematic germplasm evaluation trial was begun in 1994 using nine selected germplasm accessions (Moorock, Maliboda, Clovis, St. Annes, Kasagala, Debarayaya, Margaret, Ambakelle special and Ambakelle tall) that were already in ex situ genebanks (Perera and Peries 1993). The seedlings were planted in a completely randomized design, consisting of 15 palms per accession and five progenies per palm, representing half-sib families from each palm (15 x 5 = 75).
Because the interaction between genotype and environment on the important quantitative characters of coconut appears to be reasonably high, the genetic variation could be masked by environmental factors, making evaluation of genotypes based on phenotype inconclusive. As a result, methods of evaluating material at the molecular level are regarded to be more accurate. Restriction fragment length polymorphism (RFLP) and isozyme analysis of coconut have been given priority in the Sri Lankan research programmes.
Protocols to extract leaf isozymes, which were unsuccessful in earlier studies (Carlos 1980; Meunier 1992) because of polyphenol interferences, have been established. Recent studies have shown that the level of polymorphism shown by leaf isozymes is high compared to levels in haustorium tissues. A high level of polymorphism was detected at the esterase loci among the individuals of improved tall and tall x tall (CRIC 60), compared to the ordinary tall known as plus palm [selected mother palms), suggesting that differentiation among populations based on allelic composition may be possible (Fernando 1995).
Utilization
The analysis of the vegetative data of the progeny of selected crosses at ISG has shown that genetic gain due to crossing is very low, i.e. H2b was 0.15 (Fernando 1995). Thus, response to further selection may be limited in the highly selected pool at ISG. Hence, it was important to test the potential combining ability of seed palms at ISG with different pollen sources in order to broaden the genetic base and to test the drought tolerance and the yield potential of a cross. Four germplasm accessions, Moorock, Debarayaya, Kasagala and St. Annes, which were shown to be distinctive by multivariate discriminant analysis of the accession data, have been crossed to a selected pool of proven high and stable yielders at the ISG; at Ambakelle (Fernando and Perera 19C)3). The seedlings are now under nursery evaluation and are expected to be planted in late 1995 in multilocation trials, especially in areas where droughts are frequent.
Future work
Collecting and conserving more populations from areas which are still unexplored within the country has been proposed. Evaluation and characterization will be done in the field as well as in the laboratory using molecular methods (isozymes and RFLPs). Physiologically adapted material will be identified and conserved in situ. Accessions will be tagged for important characters such as nut yield, weight of husked nut, adaptability and stability so that they can be used in breeding programmes to develop varieties with high yields and stability for different localities in the country.
References
Carlos, C.B. 1980. Biochemical status of Cocos nucifera L., MSc thesis University of Birmingham, UK.
Carpio, C.B. 1990. Collecting strategies. Training manual of IBPGR short-term training course on collection, conservation and characterization of coconut genetic resources, Philippines. 25 November - 9 December 1991.
CDA.1993. Sri Lanka coconut statistics. Coconut Development Authority
Foale, M.A. 1991. Coconut Genetic Diversity: Present knowledge and future research needs. Proceedings of the IBPGR workshop on coconut genetic resources, Cipanas, Indonesia. 8-11 October 1991. p 4653.
Fernando, W.M.U. 1992. Report of the Genetics and Plant Breeding Division. In Report for 1992, Coconut Research institute, Sri Lanka.
Fernando, W.M.U. 1995. Unpublished data.
Fernando, W.M.U. and L. Perera. 1993. Report of the Genetics and Plant Breeding Division. In Report for 1993, Coconut Research institute, Sri Lanka.
Lawrence, M.J. and N. Rajanaidu. 1985. The genetic structure of natural populations and sampling strategy. Proceedings of the international workshop on oil palm germplasm and utilization, Selangor, Malaysia. 26-27 March, 1985. pp. 15-26.
Liyanage, D.V. 1956. Intra specific hybrids in coconuts. Bulletin No. 7. Coconut Research institute, Ceylon.
Liyanage, D.V. 1957. Annual Report of Coconut Research institute of Sri Lanka for 1957.
Liyanage, D.V. 1958. Varieties and forms of the coconut palm grown in Ceylon. Ceylon Coconut Q. 9:1-10.
Liyanage, D.V. 1961a. The use of isolated seed gardens for coconut seed production. Ceylon Coconut Q. 12:245-252.
Liyanage, D.V. 1961b. Annual Report of Coconut Research institute of Sri Lanka for 1961.
Meunier, J. 1962. Genetic diversity in coconut isozyme electrophoresis. In Proc. of the IBPGR workshop on coconut genetic resources Cipanas, Indonesia 8-11 October 1992. International Crop Network series & IBPGR, Rome 1992.
Perera, L. and R.R.S. Peries. 1993. Report of the Genetics and Plant Breeding Division. In Report for 1993, Coconut Research institute, Sri Lanka.
Peries, R.R.A. 1991. Report of the Genetics and Plant Breeding Division. In Report for 1991, Coconut Research institute, Sri Lanka.
Peries, R.R.A. 1993. Improving the efficiency of the low input coconut agro-ecology, through conservation and utilization of biodiversity. Proceedings of the 10th IFOAM conference, LincoIn University, New Zealand 1]-16 December, 1994. p. 122.
Wickramaratne, M.R.T. 1984. Report of the Genetics and Plant Breeding Division. In Report for 1984, Coconut Research institute, Sri Lanka.
Wickramaratne, M.R.T. 1987. Report of the Genetics and Plant Breeding Division. In Report for 1987, Coconut Research institute, Sri Lanka.
Zobel, B. and J. Talbert. 1984. Applied Forest Tree Improvement. John Wiley and Sons, inc., USA.
Résumé
Conservation de la biodiversité de la noix de coco (Cocos nucifera L.) à Sri Lanka
Le présent article examine les progrès de la recherche concernant la conservation de la biodiversité de la noix de coco a Sri Lanka et l'utilisation du matériel génétique pour la mise au point de cultivars améliores. Les récoltes ont été faites a la fois au hasard et systématiquement, mais la préférence a été donnée a la récolte de populations présentant des caractéristiques d'importance économique. Au total, 39 accessions ont été récoltées jusqu'ici et des doubles vent conserves dans des barques de gènes ex situ. L'examen de l'adaptation physiologique a long terme des cocotiers a conduit a la création de barques de gènes in situ en vue de produire du matériel végétal adapte aux systèmes d'agriculture durable à faible apport d'intrants. Sont également étudiées des procédures pour quantifier la variabilité entre les populations et a l'intérieur d'une population de noix de coco a l'aide de méthodes de variation quantitative et de techniques moléculaires.
Resumen
Conservación de la biodiversidad del coco (Cocos nucifera L.) en Sri Lanka
Se pasa revista en este trabajo a los avances realizados en la investigación pare la conservación de la biodiversidad del coco en Sri Lanka y el empleo de germoplasma pare el desarrollo de cultivares mejorados. Se han hecho colecciones tanto aleatorias como sesgadas aunque se dio preferencia a la recogida de poblaciones que mostraran características de importancia económica. Se han recogido en total 39 ejemplares hasta la fecha, que se han conservado por duplicado en bancos de genes ex situ. El estudio de la adaptación fisiológica a largo plazo de las palmas dio lugar al desarrollo de bancos de genes in situ pare producir material de siembra adecuado a los sistemas de un agricultura sostenible a base de bajos insumos. También se están investigando los procedimientos pare cuantificar la variabilidad entre poblaciones y dentro de ellas en el sector del coco, empleando pare ello métodos de variación cuantitativa y moleculares.
Didier Balma1, Claude. André St-Pierre2, Jean Collins and Jean Guy Parent3
1 institut d'Etudes et de Recherches Agricoles, 01 BP 476 01, Ouagadougou, Burkina
Faso
2 Département de Phytologie, Faculté des Sciences de l'Agriculture et de l'Alimentation, Université Laval, Sainte-Foy, Québec, Canada G1K 7P4
3 Service de Phytotechnie de Québec, Complexe Scientifique 2700, rue Einsten, Sainte-Foy, Quebéc, Canada G1P 3W8
Summary
Sixty landraces of pearl millet (Pennisetum glaucum (L.)) representing genetic variability of eight countries of West Africa, including Cameroon, were used to compare the amplitude of genetic polymorphism. Original seed samples and fresh leaf samples harvested from the greenhouse were screened for _- and _-esterase by polyacrylamide gel electrophoresis. For all 60 accessions, seed samples exhibited more banding patterns than leaf material. The number of isozyme bands ranged from a minimum of 17 to a maximum of 24. We observed four zones as 0 to 2, 2 to 5, 8 to 12 and 1 to 4 isozyme profiles named EST 1, EST 2, EST 3 and EST 4, respectively. Principal component analysis and similarity index studies revealed a total absence of esterase enzymatic variability or small differences among landraces, mostly those collected in eastern and south-central Burkina Faso. The similarity index average ranged from 90 to 100%. Evidence is presented to support the hypothesis that the semi- and late-maturing genotypes collected in areas with greater amounts of rainfall could arise from the same complex of cultivated populations. These populations acquired regional adaptation during their evolution after several modifications imposed by the limitation of the geneflow which could have occurred only among earlier genotypes. However, large esterase enzymatic variability was found among earlier landraces collected from northern and central Burkina Faso, including those of the other countries of West Africa. These results, on one hand, showed the possible different sources of enzymes from the wild progeny through the West Africa region and, on the other hand, confirmed the possibility of different origins of the species' domestication. However, the technique has potential for practical application; subsequent studies should refine the procedure and discover other pearl millet enzyme systems based on corresponding allele number that might reveal desirable agronomic characters, useful parent genotypes for breeding programmes, or samples to be conserved in genebanks.
Introduction
Interest in collecting the genetic diversity available in cultivated pearl millet, its wild relatives and wild species began 20 years ago. Several institutes started to evaluate the genetic diversity of these collections. These evaluations sometimes aimed at examining the biological structure of the species, the information on elements of their evolution and their domestication. Now, the process will have to evolve toward the use of the genetic resources by plant breeders and rationalization of the management of the genebanks to minimize the conservation of redundant samples, which is sometimes useless and costly.
Genetic polymorphism is based on a certain number of genes coding for 8-12 enzymes, including the _-esterase (EST) which has already been obtained in pearl millet. Bono (1973) showed the similarity between malian and burkinabe pearl millets, and Tostain et al. (1987) showed the great diversity between Nigerian pearl millets from eight enzymes studied seed to seed and separated on starch gel and whose _-esterases were separated on polyacrylamide. According to Sarr et al. (1988), the relative efficiency of each pollen source is determined by gene-specific alleles coding for isozymes as well as for esterase and alcohol dehydrogenase.
Esterase activity can be measured easily from extracts of plants. The esterase zymogrammes contain a characteristic pattern of a large number of strips (sometimes more than ten) controlled by several loci; these esterase isozymes that are distinct molecular models of only one enzyme acting on the _-naphthylacetate and the _-naphthylacetate are esterase carboxyls (EC 3.1.1.1) (Cubadda and Quattrucci 1974). They have been used often to study the intra- and inter-specific polymorphism (Brown and Weir 1983). One of the advantages of using isozymes to study polymorphism is that a range of enzyme loci on one individual can be studied easily using a small quantity of material with minimum preparation and cost (Kidambi et al. 1990).
Our objective was to use electrophoresis to determine, rapidly and inexpensively, the existence of genetic polymorphism in esterase enzymes of 60 ecotypes of Pennisetum glaucum species collected in eight countries of West Africa, including Cameron
Materials and methods
Sixty ecotypes selected at random from the 110 ecotypes used for the evaluation of agromorphological characters were used (Table 1). The seeds of pearl millet collected in West Africa (Fig. 1) derive directly from the genebank in Ottawa, Canada. Ground seeds and leaves of young plants cultivated in pots in greenhouses were used.
Table 1. List of ecotypes studied
No. |
CIN1 |
Origin2 |
No. |
CIN1 |
Origin2 |
No. |
CIN1 |
Origin2 |
1 |
50 |
Nigeria |
21 |
2851 |
CBF |
41 |
3209 |
NBF |
2 |
87 |
Nigeria |
22 |
2861 |
CBF |
42 |
3669 |
EBF |
3 |
724 |
Cameroon |
23 |
2864 |
CBF |
43 |
3671 |
EBF |
4 |
733 |
Cameroon |
24 |
3141 |
CBF |
44 |
3672 |
EBF |
5 |
1968 |
Mali |
25 |
3144 |
NBF |
45 |
3674 |
EBF |
6 |
1114 |
Mali |
26 |
3145 |
NBF |
46 |
3676 |
EBF |
7 |
1273 |
Senegal |
27 |
3152 |
NBF |
47 |
3679 |
EBF |
8 |
1309 |
Senegal |
28 |
3156 |
NBF |
48 |
36B1 |
EBF |
9 |
1501 |
Niger |
29 |
3157 |
NBF |
49 |
3683 |
EBF |
10 |
1510 |
Niger |
30 |
316B |
NBF |
50 |
3687 |
EBF |
11 |
2001 |
Niger |
31 |
3176 |
NBF |
51 |
3690 |
EBF |
12 |
2034 |
Benin |
32 |
3180 |
NBF |
52 |
3693 |
EBF |
13 |
2812 |
CBF |
33 |
3183 |
NBF |
53 |
3699 |
EBF |
14 |
2816 |
CBF |
34 |
3187 |
NBF |
54 |
3702 |
SCBF |
15 |
2821 |
CBF |
35 |
3193 |
NBF |
55 |
3705 |
SCBF |
16 |
2830 |
CBF |
36 |
3201 |
NBF |
56 |
3708 |
SCBF |
17 |
2837 |
CBF |
37 |
3202 |
NBF |
57 |
3710 |
SCBF |
18 |
2841 |
CBF |
38 |
3204 |
NBF |
58 |
3716 |
SCBF |
19 |
2844 |
CBF |
39 |
3206 |
NBF |
59 |
3751 |
SCBF |
20 |
2846 |
CBF |
40 |
3207 |
NBF |
60 |
3769 |
SCBF |
1 CIN=Collector identification number assigned to samples stored in genebank at Ottawa.
2 CBF=central Burkina Faso; NBF=northern Burkina Faso; EBF=eastern Burkina Faso; SCBF=south-central Burkina Faso
- Fig. 1. Collecting area of pearl millet in countries of West Africa
In a mortar on ice, 0.35 g of seeds (about 45 seeds on average) was ground with 1.2 ml of extraction stopper [Tris(hydroxymethyl)aminomethane (0.1M, pH 7.5)] containing 1 mM sodium ethylenediaminetetraacetate (EDTA), 10 mM magnesium chloride, 10 mM potassium chloride, 100 mg/ml polyvinyl pyrrolidone of 40 000 average molar mass (Cousineau and Donnelly 1989) and 0.1 % 2-mercaptoethanol (Gottlieb 1981). The paste was made homogeneous and then decanted into a 1.5-ml conical tube and centrifuged for at least 15 min at 13 000g. The liquid floating on the surface was recovered and transferred into another conical tube which was stored at 70°C until used. The purpose of making the paste homogeneous was to liberate soluble proteins and preserve them in solution.
Green leaves (0.30 g) newly collected from ten plants 24 weeks old were introduced into a conical tube on ice. Silica carbide (carborundum, grains 320) was added to facilitate grinding. The tissues were ground with a glass rod that just fit into the conical tube. The centrifugation, recovery and conservation of liquid were similar to those for the seeds.
Thirty minutes before prefocalization, the tubes containing the liquid were placed on a mixture of water and ice (0°C) to thaw. We have since used a Fisherbiotech apparatus, FB 1001, to thaw the extracts. The apparatus was pre-cooled to 10°C; electricity was controlled by a 2197 LKB Bromma regulator.
A LKB polyacrylamide gel with dimensions of 245 x 110 x 1 mm was cut to 205 x 110 x 1 mm and placed on a rest previously moistened with kerosene to favour thermic conductivity. Electrolysis bands LKB 280 x 6 x 1 mm carefully cut to the gel dimension were dipped in anodic (0.5M acetic acid) or cathodic (0.5M NaOH) solutions. After being lightly sponged, they were put in contact with the electrodes on the gel. The apparatus was set to function for 15 min of prefocalization at a contact current of 20 out 200 V.
On a prefocalized gel, a silicon applier pierced with holes every 2 mm was placed 1 cm from the anode and the samples (2.5 ml) were put into the holes. The migration was conducted at a constant power of 11 W allowing free fluctuation of the electric tension. After 30 min. the silicon applier was removed and the excess liquid carefully sponged away. Then the electrophoresis continued for 2 hours.
After electrophoresis the gel was put into a specific colouring, consisting of 50 ml phosphate solution (0.2 M, pH 6.4), 50 mg _-naphthylacetate and 50 mg _-naphthylacetate in 5 ml acetone, 100 mg of Flast Blue RR salt (hemisalt of zinc chloride benzoylamino-4-dimethoxy-2.5-benzendiazonium), according to the method of Cheliak and Pitel (1985). This solution was incubated with shaking in darkness at 37°h for 2 h.
The gel was transferred into decolouration solution (26% methanol and 7% acetic acid) taking care to keep it in the dark. For drying, the gel was rinsed with distilled water and dipped into 10% glycerol for 30-45 minutes. It was then covered with a plastic sheet previously dipped into the glycerol solution to prevent air pockets, and left in darkness at room temperature, protected from dust, for at least 3 days.
Each band of isozyme activity was identified with a capital letter going in alphabetical order following the isoelectric point from the lowest to the highest. The numeric values 0, 1, 2 and 3 correspond respectively to the absence or the presence, on a semiquantitative scale, of the visually observed band intensity: 0 = absent, 1 = slightly coloured, 2 = fairly coloured, 3 = very coloured.
The similarity between various profiles was expressed by the similarity index. The percent similarity was calculated for the different ecotypes following the procedure of Whitney et al. (1968):
Percent similarity = [(Number of pairs of similar bands) / (Total number of bands in both ecotypes)] x 100
Results and discussion
The difference between the enzymatic profiles observed in the ecotypes concerns the presence or the absence, the intensity of the bands and their isoelectric position (Figs. 2 and 3). The symbol used by Kahler and Allard (1970) permits identification of the esterase band according to its position on the zymogramme. It was possible to identify in the 60 populations a total of 24 bands that are very distinct at pH 4 0 6 4 These bands of esterase isozymes are named in alphabetical order A to X. The total number of isozyme bands varies between 17 and 24. Cubadda and Quattrucci (1974) enumerated 17 in barley. Isozymes A, B. M and X are common to the 60 ecotypes (Fig. 4). With the allogame species any population can often be representative of the whole species because of a more frequent allele for a given gene as Gottlieb (1981) indicates for the subspecies Staphnomeria exigua and as Kidambi et al. (1990) notes for the species Onobrychis.
- Fig. 2. Isozyme banding patterns obtained with ground seeds from a single ecotype
Fig. 3. Isozyme banding patterns obtained with individual seed from ecotype 3141
The 24 bands are distributed in four zones named EST 1, EST 2, EST 3 and EST 4. There are 2 isozymes in EST 1, 26 in EST 2, 8-12 in EST 3 and 1-4 isozymes in EST 4. As to the intensity of the bands or the designation of the different esterase zones, we are reluctant to comment because of the lack of measurement of the quantities or the enzymatic activities.
In the zone of EST 1, there are two bands for all ecotypes. Band B is more voluminous and more intensely coloured than band A in almost all of the ecotypes.
The EST 2 zone includes isozyme bands of different relative intensities and the number varies between 2 (ecotype 2851) and 5. The isozymes E, F. G and H are present in all ecotypes except in ecotype 2851. They are particularly coloured in ecotype 2864. Bands C and D are very intense for ecotypes 2830, 3687, 724, 733, 3702 and 3710.
The zone of EST 3 includes the most active part whose number of bands varies between eight (2851) and 12. For all ecotypes except ecotypes 1273, 2861, 1501 and 3699, the most intense band is L. The number of bands varies mainly in foreign ecotypes and in those from northern Burkina Faso.
Though less complex than the EST 3 zone, the EST 4 zone possesses some very intensely coloured bands and a great variability. The number of bands varies between one and four. The isozymes T. V and W respectively are present at only 13%, 4% and 67%. The band T. present in most ecotypes, is a very intense pink colour, except in 724, 733, 1114, 2837, 2844, 3152, 3176, 3202, 3207, 3209, 3683, 3690, 3699, 3710 and 3769 where it is also smaller. Trigui (1984), Tostain and Riandey (1984) and Trigui et al. (1986) showed that for esterases separated on starch gel, some bands are differently coloured in grey by anaphthylacetate or in pink by B-naphthylacetate. They reported that five bands of esterase are observed in leaves with a- or p-naphthylacetate as a substratum and that of the pluri-allelic loci, the most outstanding would be the locus EST E possessing 7 alleles.
- Figure. 1, Programmes showing the banding patterns of 60 pearl millet ecotypes for esterase. Isozymes 2, 3-5, 8-12 and 1-4 are in EST 1, EST 2, EST 3 and EST 4 respectively.
Figure 4 shows the main characteristics of the electrophoresis profiles. The variability expressed in the EST 4 zone is by far the most important, followed by zone EST 3.
We previously showed the discrimination produced by the enzymatic esterase system among the 60 ecotypes. In general, the esterase system showed less polymorphism within the 60 ecotypes when we used several seeds or several leaves at the same time. The observation of three ecotypes with a particular colour and band position showed a very high intra-ecotype variability, that is ecotypes 50, 2830 and 3141 (Fig. 3).
Figure 5 gives the results of the principal component analysis (PCA) of the isozyme banding patterns. The four main axes that we have not reported here account for 26, 20, 16 and 15% respectively, giving a cumulated total of 77% of variance, demonstrating the association among several alleles (Tostain et al. 1987) and the presence of strongly associated groups. This analysis is based mainly on the presence or the absence of bands. As such it is phenotypical and its interpretation will not take into account genetic control.
- Fig. 5. Graphic representation of the principal component analysis according to the number of isozyme banding patterns.
The PCA graph shows that there are three ecotype groups which can present a certain phenotypical diversity on the plan 1 - 2 for the extracts of seeds: these consist of foreign ecotypes and of those of northern and central Burkina Faso. The point of maximum diversity would be in the middle of the group of the ecotypes of central Burkina Faso: this indicates that the greatest genetic diversity would be within the ecotypes as Tostain et al. (1987) stipulate. They show the existence of the greatest diversity within the ecotypes of the north (high Sahelian zone) in relation to the centres of origin of the species. Our results do not contradict those of Tostain et al. (1987), because as drought advances toward the southern Sahara, the local farmers tend to sow early ecotypes. That can explain the increasingly rapid displacement of the ecotypes of the north toward the central and southern parts of the country. The further we go toward the zone where semi-late and late varieties are cultivated, the more the enzymatic variability is reduced.
To estimate the variation of the bands of esterase isozymes between the different ecotypes studied, we calculated the percentage of similarity between the ecotypes of the same area of prospection, between 60 ecotypes, then between ecotypes known to be different between themselves. For the ecotypes collected from Burkina Faso: from the centre of the country, the similarity indication varies between 53% and 100% with averages as high as 71.8% and 93.7%; for those from the north, the east and the south central, the similarity indication varies individually between 82.0 and 100% with averages between 83% (3176) and 99.2%. Only the ecotypes of the central region are all different from one another. Ecotype 2851 was the most distinct, having only 53% of similarity. The graphic representation of these ecotypes on the PCA plan 1 - 2 gave a great distribution of those of central Burkina Faso. The discriminating esterase systems show that:
· two of the northern ecotypes, 3145 (93.4%) and 3176 (91.7%) have average similarity indications less than 95%, which gives two distinct ecotypes plus an outcrossed population constituted by the remaining ecotypes of the same area of prospection;
· of the ecotypes of eastern region, only the ecotype 3699 (92.9%) possesses an average less than 95%: we keep from this group the distinct ecotype and an outcrossed population constituted by the rest of the ecotypes
Table 2. Similarity indices for esterases in 23 ecotypes collected from Burkina Faso and from other countries of West Africa (in %)
Ecotype |
87 |
724 |
733 |
1068 |
1114 |
1273 |
1309 |
1501 |
1510 |
2001 |
2034 |
2812 |
2837 |
2851 |
2861 |
2864 |
3141 |
3145 |
3176 |
3204 |
3699 |
3769 |
Mean |
50 |
88 |
90 |
90 |
93 |
95 |
90 |
90 |
93 |
88 |
84 |
93 |
95 |
93 |
78 |
79 |
90 |
93 |
93 |
91 |
95 |
88 |
77 |
89.4 |
87 |
- |
97 |
95 |
90 |
97 |
97 |
97 |
95 |
87 |
68 |
85 |
86 |
84 |
73 |
74 |
87 |
90 |
90 |
88 |
95 |
84 |
89 |
88.0 |
724 |
- |
100 |
93 |
95 |
95 |
95 |
97 |
90 |
72 |
88 |
88 |
87 |
71 |
78 |
90 |
93 |
93 |
90 |
98 |
87 |
92 |
89.9 |
|
733 |
- |
93 |
95 |
95 |
95 |
97 |
97 |
72 |
88 |
88 |
87 |
71 |
78 |
90 |
93 |
93 |
90 |
98 |
84 |
92 |
93.4 |
||
1068 |
- |
93 |
88 |
88 |
90 |
98 |
81 |
95 |
100 |
95 |
74 |
86 |
93 |
100 |
95 |
93 |
95 |
90 |
89 |
91.5 |
|||
1114 |
- |
90 |
95 |
93 |
90 |
79 |
93 |
93 |
88 |
78 |
79 |
95 |
93 |
95 |
91 |
98 |
88 |
87 |
90.9 |
||||
1273 |
- |
95 |
92 |
85 |
67 |
83 |
88 |
87 |
65 |
78 |
85 |
93 |
88 |
90 |
93 |
87 |
86 |
87.1 |
|||||
1309 |
- |
92 |
85 |
72 |
88 |
83 |
82 |
71 |
72 |
90 |
88 |
88 |
86 |
93 |
82 |
86 |
86.9 |
||||||
1501 |
- |
92 |
74 |
85 |
86 |
89 |
73 |
74 |
87 |
90 |
90 |
88 |
95 |
84 |
89 |
88.4 |
|||||||
1510 |
- |
83 |
98 |
93 |
92 |
71 |
83 |
90 |
98 |
93 |
90 |
93 |
87 |
79 |
88.9 |
||||||||
2001 |
- |
86 |
82 |
80 |
87 |
75 |
78 |
81 |
77 |
74 |
76 |
69 |
76 |
76.9 |
|||||||||
2034 |
- |
93 |
90 |
74 |
81 |
80 |
95 |
91 |
88 |
98 |
85 |
82 |
88.1 |
||||||||||
2812 |
- |
90 |
76 |
82 |
93 |
95 |
96 |
93 |
95 |
86 |
83 |
89.3 |
|||||||||||
2837 |
- |
79 |
86 |
87 |
95 |
90 |
88 |
90 |
84 |
80 |
87.4 |
||||||||||||
2851 |
- |
53 |
76 |
74 |
70 |
67 |
74 |
73 |
62 |
72.3 |
|||||||||||||
2861 |
- |
93 |
86 |
82 |
84 |
86 |
74 |
76 |
79.0 |
||||||||||||||
2864 |
- |
93 |
88 |
86 |
93 |
82 |
79 |
87.5 |
|||||||||||||||
3141 |
- |
95 |
93 |
95 |
90 |
87 |
91.4 |
||||||||||||||||
3145 |
- |
98 |
95 |
90 |
87 |
89.9 |
|||||||||||||||||
3176 |
- |
93 |
88 |
80 |
87.7 |
||||||||||||||||||
3204 |
- |
90 |
82 |
91.8 |
|||||||||||||||||||
3699 |
- |
86 |
84.4 |
||||||||||||||||||||
3769 |
- |
83 |
· among the south central ecotypes, there is only the ecotype 3751 which possesses an average of 83%: there would be therefore this sample and the rest to constitute a distinct ecotype and a new outcrossed population.
For all ecotypes of Burkina Faso, we can, without any extrapolation of these results, envisage the following restructuring:
· eleven ecotypes of central Burkina Faso;
· three ecotypes and two new samples as outcrossed populations;
· one ecotype and a new outcrossed population in eastern Burkina Faso, as well as in the south central region. There may be a total of 19 samples instead of 48 (31 %) to be conserved in the genebank.
The indications of similarity for all ecotypes considered to be different are reported in Table 2: the ecotypes of Burkina Faso that have an indication of similarity less than 95% are compared with those of foreign countries. The values vary from 62 to 100% with averages between 72.3 and 91.8%. Ecotype 2851 has the lowest average (72.3%) followed by 2001 (76.9%).
In all the populations studied there are two esterase enzyme groups with great variability: the group of foreign ecotypes and the group of central Burkina Faso ecotypes. The other groups have a minor enzyme diversity. There is some similarity between isozymatic, morphological and geographical distributions.
The ecotypes of south central and eastern Burkina Faso constitute the majority of homogeneous groups and enzymatically are much less variable. The diversity observed among foreign ecotypes is not surprising because the geographical distances between the countries are sufficiently large to produce these differences in pearl millet. However, it is difficult to explain the net amount of diversity among the ecotypes of the north and the remaining ecotypes of the other regions of Burkina Faso by geographical variations. Nevertheless, it is known that the ecotypes of the south central and the eastern region are semi-late genotypes that are sensitive to the photoperiod. The ecotypes of the central region would constitute a transitory zone toward the north, and would therefore have a greater diversity; but this diversity is not very visible on the enzymatic plan. This cannot be explained by environmental factors.
The ecotypes of the northern region could come from several domestications and from several wild parents growing along the southern Sahara (Tostain et al. 1987). The diversity among the ecotypes of the north could be explained by their proximity to the Nigerian pearl millets known as the most varied of the Sahelian region, and thus would have benefited from a flow of genetic exchanges. These ecotypes are also at the crossroads of the domestication centres recommended by Harlan (1971).
The semi-late and the late varieties, which are probably the result of a migration of early varieties from the north to the south, could have become more specialized to conform to a rainy and regular climate. However, the populations of early pearl millets could also be the consequence of a long migration process through arid and even rainy regions of West Africa.
The discordance of esterasic distances comes probably from the fact that enzymes remain neutral in the interaction between genotype and environment. The differences observed at the level of the esterase showed that out of 60 ecotypes studied, 30 can be considered similar. To complete the morphological data, redundant samples can be detected. For Fraleigh (1989), grouping samples according to genetic resemblance allows the constitution of a reduced representative collection (idea of core collection). The samples that form the groups can be maintained together in a strategy of 'dynamic conservation' (Pernès et al. 1984). These sources of information are not only necessary for the rationalization of the conservation of the genetic resources, but they are also a means for breeders to have at their disposal the available genetic variability. Though the observation of esterasic similarity indicates great genetic proximity among the majority of the ecotypes collected, the isozyme bands present sufficiently distinct profiles to permit ecotype differentiation, and the intensity of the bands can assist in their identification.
Conclusion
These results show that genetic evaluation of allogame populations requires the study of phenotypical and enzymatic variations to optimize genetic diversity of the cultivated forms. The revelation of genetic determinism of esterase isozymes and the use of electrophoresis to analyze other pearl millet enzymes could permit better understanding of the genetic variability of the cultivated species. However, this method needs to be developed so that the genetic control of isozymes can be explained. In support of multivariate analysis methods, enzymatic discrimination permits visualization of the formation of new outcrossed populations even genetically richer and less redundant for germplasm conservation. However, the genetic scope of this phenotypical diversity must be checked by studying the crossing between the collected and already evaluated ecotypes, in particular enzymatic polymorphism and restriction fragment length polymorphism. It is also necessary to find the cause of the regional distribution of the diversity of the ecotypes of West Africa. That knowledge will shed light on the evolution of pearl millet species, the content of the genetic richness that will permit improvment of the prospection techniques, the identification and the sampling of the ecotypes in their primary and secondary differentiation areas in order to have better strategies of collecting and of conservation of genetic resources in these regions of Africa.
References
Bono, M.1973. Contribution à la morphosystématique des Pennisetum annuels cultivés pour leurs grains en Afrique occidentale francophone. Agron. Trop (Paris) 28:229-356.
Brown, A.H.D. and B.S. Weir. 1983. Measuring genetic variability in plant populations. Pp. 219-239 in Isozymes in plant genetics and breeding, Part A. (S.D. Tanksley and T.J. Orton, eds.). Elsevier, Amsterdam, Oxford.
Cheliak, W.M. et J.A. Pitel. 1985. Techniques d'électrophoresè sur gel d'amidon des enzymes d'essences d'arbres forestiers. Institut forestier national de Petawawa, service canadien des forêts. Rapport d'information PI - X - 42F, 47 pp.
Cousineau, J.C. and D.J. Donnelly. 1989. Identification of raspberry cultivars in vivo and in vitro using isoenzyme analysis. Hort. Sci. 24:490-492.
Cubadda, R. and E. Quattrucci. 1974. Separation by gel electrophocusing and characterization of wheat esterases. J. Sci. Food Agric. 25:417-422.
Fraleigh, B. 1989. Importance des barques de resources phytogénétiques. Amélioration et Protection des Plantes Vivrières Tropicales (Claude André Saint-Pierre et Yves Demarly Resp. Sci.). Ed. AUPELF-UREF. John Libbey Eurotext. Paris. pp. 13-18.
Gottlieb, L.D. 1981. Electrophoretic evidence and plant populations. Prog. Phytochem. 7:1-46.
Harlan, J.R. 1971. Agricultural origins: Centers and non-centers. Science 14:468-474.
Kahler, A.L. and R.W. Allard. 1970. Genetics of isozyme variants in barley. I. Esterases. Crop Sci. 10:444-448.
Kidambi, S.P., J.R. Mahan, A.G. Matches, J.J. Burke and R.R. Nunna. 1990. Genetic variability for esterase enzyme in Onobrychis species. Theor. Appl. Genet. 80:433436.
Pernes J. et al., 1984. Gestion des ressources génétiques des plantes: Tome I et Tome 11. Monographies - Agence de Coopération Culturelle et Technique, Paris.
Sarr, A., M. Sandmeier and J. Pernes. 1988. Gametophytic competition in pearl millet, Pennisetum typhoides (Stapf et Hubb.). Genome 3:924-929.
Tostain, S. et M.F. Riandey. 1984. Polymorphysme et déterminisme génétique des enzymes de mil (Pennisetum glaucum L.). Etude des alcools deshydrogénases, catalases, endopeptidases et estérases. L'Agron. Trop. 39(4):335-345.
Tostain, S., M.F. Riandey and L. Marchais. 1987. Enzyme diversity in pearl millet (Pennisetum glaucum). Theor. Appl. Genet. 74:188-193.
Trigui, N.1984. La variabilité génétique des mils (Pennisetum typhoides (Bur.) Stapf & Hubb.) de Tunisie. Etude biométrique et analyse du polymorphisme enzymatique. Thèse 3 e cycle (Developpement et amélioration des végétaux), Univ. Paris Xl - Orsay, 125 p.
Trigui, N., M. Sandmeier, M. Salanoubat et J. Pernes.1986. Utilisation des données enzymatiques et morphologiques pour l'etude des populations et de la domestication des plantes: I. Séparation et identification génétique d'isozymes chez le mil (Pennisetum typhoides Bur. Stapf & Hubb.).
Whitney, P. J., J.G. Vanghum and J.B. Heale. 1968. A disc electrophoretic study of the proteins of Verticillium albo-atrum, V. dahliae and Fusarium oxysporium with reference to their taxonomy. J. Exp. Bot. 19:415-426.
Résumé
Méthode d'identification du polymorphisme génétique des enzymes de l'estérase dans les écotypes de Pennisetum glaucum.
Soixante variétés locales de mil chandelle (Pennisetum glaucum (L.)) représentant la variabilité génétique de huit pays d'Afrique de l'Ouest, dont le Cameroun, ont été utilisées pour comparer l'amplitude du polymorphisme génétique. On a testé des échantillons de semences originaux et des échantillons de feuilles fraîches récoltes en pépinière pour détecter l'estérase a et _ par electrophorese en gel de polyacrylamide. Pour les 60 obtentions, les échantillons de semences ont affiche davantage de spectres de bande que le matériel feuillu. Le nombre de bandes d'isozymes allait d'un minimum de 17 à un maximum de 24. Nous avons observe quatre zones: 0 3 2, 2 a 5, 8 à 12, et 1 à 4 profils d'isozymes dénommes respectivement EST 1, EST 2, EST à et EST 4. L'analyse en principales composantes et des études de l'indice de similitude ont révère une absence totale de variabilité dans les enzymes de l'esterase ou de petites différences entre variétés locales, principalement celles récoltées dans l'est et le centre-sud du Burkina Faso. La moyenne de l'indice de similitude allait de 90 a 100 %. On a démontre que les génotypes 3 maturation intermédiaire et tardive récoltes dans des zones a plus forte pluviosité pourraient venir du même ensemble de populations cultivées. Ces populations se vent adaptes aux conditions locales durant leur évolution apres plusieurs modifications imposées par la limitation du flux de gènes qui pourrait s'y être produite seulement parmi les premiers génotypes. Toutefois, une forte variabilité te des enzymes de l'estérase a été relevée parmi les premières variétés locales récoltées dans le nord et le centre du Burkina Faso, y compris celles des autres pays d'Afrique de l'Ouest. Ces résultats ont d'une part montre les différentes sources possibles des enzymes provenant de descendants sauvages en Afrique de l'Ouest et ont d'autre part confirme la possibilité que la domestication de l'espèce ait diverges origines. Néanmoins, la technique pourrait avoir des applications pratiques; des études complémentaires devraient peaufiner la procédure et découvrir d'autres systèmes d'enzymes du mil chandelle fondes sur le nombre d'allèles correspondent qui pourraient révéler des caractères agronomiques souhaitables, des génotypes directement utilisables comme parents pour les programmes d'amélioration ou des échantillons 3 conserver dans des barques de gènes.
Resumen
Método para identificar el polimorfismo genético de las enzimas de esterasa en ecotipos de Pennisetum glaucum.
Para comparar la amplitud del polimorfismo genético se utilizaron 60 variedades de mijo perla (Pennisetum glaucum (L.)), que representaban la variabilidad genética de ocho países de Africa occidental, incluido Camerún. Para la esterasa alfa y beta se seleccionaron, mediante electrofóresis del gel de poliacrilamida, muestras de semillas originales y de hojas verdes recogidas en invernadero. Para todos los 60 ejemplares, las muestras de semillas mostraron mas pautas de distribución en bandas que el material foliar. El numero de bandas isozimas oscilaban de un minimo de 17 a un máximo de 24. Observamos cuatro zonas de perfiles de isozimas de 0 a 2, 2 a 5, 8 a 12 y 1 a 4, que denominamos EST 1, EST 2, EST 3 y EST 4, respectivamente. De un análisis de los componentes principales y de estudios de los índices de similaridad, resultó una ausencia total de variabilidad en las enzimas de esterasa o pequeñas diferencias entre las especies, sobre todo en las recogidas en la zona oriental y centromeridional de Burkina Faso. La media de los índices de similaridad fluctuó del 90 a 100% Se aportan también pruebas que apoyan la hipótesis de que los genotipos de madurez tardía y mediana recogidos en zonas con mayores volúmenes d e pluviosidad ad podrían proceder del mismo conjunto de poblaciones cultivadas. Estas poblaciones adquirieron una adaptación a la región durante su evolución tras varíes modificaciones impuestas por la limitación del flujo genético que había podido darse solo entre los genotipos anteriores. No obstante, se hallo una gran variabilidad de enzimas de esterasa entre las especies anteriores recogidas de las regiones septentrional y central de Burkina Faso, inclusive en las de los otros paises de Africa occidental. Por una parse, estos resultados mostraban las posibles fuentes distintas de enzimas respecto de la progenie silvestre en toda la región de Africa occidental y, de la otra, confirmaba la posibilidad de distintos orígenes en la domesticación de las especies. Sin embargo, esta técnica encierra posibilidades de aplicación practica; en los estudios que se hagan en adelante debería perfeccionarse el procedimiento y descubrir otros sistemas de enzimas del mijo perla basados en un numero de alelos correlativos que tal vez revelen caracteres agronómicos convenientes, o aporten genotipos parentales útiles pare programas de fitogenética, o muestras pare su conservación en los bancos de genes.
Fermín de la Puente1, Daniel F. Austin2 and Jaime Díaz3
1 Departmento de Recursos Genéticos, Centro internacional de la Papa, Apartado Postal 1558, Lima, Perú
2 Department of Biological Sciences, Florida Atlantic University, Boca Raton, FL 33431, USA
3 Centro internacional de Mejoramiento de Maíz y Trigo, Apdo. Postal 6-641, Mexico 06600, D.F. Mexico
Summary
The different common names given to the sweetpotato and its cultivars in different Latin American and Caribbean countries are summarized. The origins of these names in each country were studied and found to be related to physical traits, people regions, localities and uses. In general the common names are determined by a single criterion.
Introduction
In 1985 the Centro internacional de la Papa (CIP) began to form a genebank of the sweetpotato (Ipomoea batatas (L.) Lam.), commonly known as camote, batata or boniato, with material collected in Latin America and the Caribbean. Toward this goal a number of systematic excursions were organized in different localities and habitats in Latin America and the Caribbean. These excursions did not target specific ethnic groups, but sought all people who were growing and using sweetpotatoes. Both Amerindian and non-indigenous populations were sampled. Data sheets with predetermined questions were used to obtain both ethnobotanical and horticultural data about the plants.
As of May 1992, some 16 countries had been explored, and collections made of 5395 accessions of cultivated and wild germplasm. Of the germplasm collected, some 3422 (63.4%) accessions were of the cultivated species, 171 (3.2~o) were of wild Ipomoea batatas and 1807 (33.4%) were of wild species. The cultivated germplasm was collected in Argentina, Bolivia, Brazil, Colombia, Dominican Republic, Ecuador, Honduras, Jamaica, Mexico, Nicaragua, Panama, Paraguay, Peru and Venezuela.
Within the cultivated germplasm samples there were different common names for the cultivars and cultivated strains. These names varied among the countries, states, departments, provinces, municipalities, towns and villages according to the political divisions of each country. The origins of the common names are related to harvesting and cooking customs, and to the dominant morphological characteristics of the plants and fleshy roots. This paper summarizes the broad patterns among the common names currently used.
Common names
The following common names, along with numerous orthographic variants (Austin 1988), are most frequently used for the cultivated plants in Latin American and Caribbean countries (Table 1): batata, camote, boniato, sweetpotato, palate doce, batata doce, chaco, apichu, kumara, is, chin, mabi and yety. These names are derived from local dialects and /or languages. and may have the same meanings when translated. For example, both batata and camote are used for the plants in different Spanish-speaking regions. Austin (1988) made no attempt to list all common names for the species as had Yen (1974). Instead, Austin (1988) examined linguistic relationships between the principal lineages of names, presented evidence that batata and camote are cognates, and concluded that the current use of these two names is related to historical distribution. In Cuba the same plants are called boniato by Spanish-speakers. Sweetpotato, patate dôce and batata dôce are English, French and Portuguese versions respectively of the same concept.
Table 1. Common names of sweetpotato in Latin American and Caribbean countries and possible origin of the names
Common name |
Language/origin |
Countries of usage |
Batata |
Chibchan->Spanish |
Argentina, Bolivia, Colombia, Dominican Republic, Ecuador, Nicaragua, Panama, Paraguay, Puerto Rico, Venezuela |
Camote |
Nahuatl->Spanish |
Bolivia, Chile, Costa Rica, El Salvador, Guatemala, Honduras, Mexico, Panama, Peru, Uruguay |
Boniato |
?Taino->Spanish |
Cuba, Uruguay |
Sweetpotato |
English->sweet+batata |
Jamaica, St. Vincent |
Palate dôce |
French->bataba+sweet |
Haiti |
Batata dôce |
Portuguese->bataba+sweet |
Brazil |
Chaco |
Arawakan? |
Venezuela |
Apichu |
Quechuan |
Bolivia, Peru |
Kumara |
Quechuan |
Peru |
Is, sis, chin |
Mayan |
Guatemala |
Camotli |
Nahuatl |
Mexico |
Mabi |
Cariban |
Honduras |
Kuwas |
? |
Panama |
Yety |
Tupí-Guananí |
Argentina, Paraguay |
Our research has found that many additional names are used. For example, Table 2 shows the countries that have the most diversity in common names. Some names are widely used but others are more regional (Table 3). No attempt has been made to analyze the more regional names.
Table 2. Latin American countries with the greatest diversity of common names of sweetpotato
Country |
No. of names |
Peru |
394 |
Argentina |
115 |
Ecuador |
66 |
Colombia |
58 |
Bolivia |
55 |
Honduras |
50 |
Dominican Republic |
28 |
Paraguay |
26 |
Brazil |
25 |
Table 3. Common names with the widest geographic distribution in Latin American and Caribbean countries
Common name |
Country used |
Amarillo |
Argentina, Bolivia, Colombia, Ecuador, Honduras, Mexico, Panama, Paraguay, Republic, Venezuela |
Blanco |
Argentina, Bolivia, Brazil, Colombia, Ecuador, Honduras, Mexico, Paraguay, Peru, Dominican Republic |
Colorado |
Argentina, Colombia, Ecuador, Honduras, Panama, Paraguay, Peru, Dominican Republic |
Morada |
Argentina, Bolivia, Colombia, Honduras, Mexico, Panama, Peru, Dominican Republic |
Rosado |
Argentina, Colombia, Honduras, Mexico Dominican Republic |
Camote papa |
Argentina, Ecuador, Peru, Honduras, Mexico |
Camote rosado |
Bolivia, Honduras, Peru |
Names of cultivate
Most common names may be grouped according to one predominant criterion. However, some names are more complex and are based on more than one trait. Simple names are typically based on a single criterion: morphological, physiological and culinary characteristics; production capacity; geographic origin; person who introduced the cultivar; morphological similarity with other cultivars.
Vegetative parts are the basis for some simple names. In Peru the name apichu is applied to plants with a trailing habit, and kumara to those which have a compact habit. This usage is of long-standing historical application in spite of comments to the contrary by Brand (1971). Some plants named for leaf shape include pate de pave, pie de gallina, hoja de pato, pate de gallo, pate de rana, sies pumas, and hoja redonda.
Names given because of skin colouration are: blanco, amarillo, morado, Colorado, rosado, rojo, negro and matizado. Among names based on root shapes are: blanco alagrada, colorada chica, amarilla redonda, amarilla grant and menudita. Typical names based on flesh colour are: blanco, amarillo, morado, blanco matizado an anaranjada.
Flesh colour also is the basis for names: camote used for cultivars with yellow flesh and batata for thou with white flesh. This association is continued by English speakers who call the yellow-fleshed roots 'yams' and the white-fleshed roots 'sweetpotatoes'. Common names based on cooking qualities include: yema de huevo, amarillo dulce, blanco sal, cambray dulce, morado harinoso, camote de mesa, camote coche and camote de sopa.
Names based on vegetative period (physiology) are tresmesino, noventero, cuarentón, cuarenta días and se meses.
Other names are based on production capacity. Thee include tumba burro, cargamento, rompe piedra, siete bold and amarillo rompe costal.
A few names based on geographic origin are Americana, Brasileira, Chileno, Cubano, Italiano, Japone Haitiana, Venezolana, Peruano, Serranita, Costeña; Chiclayana, Bogotana, Piurana, Trujillana, Paramonguino Mochera, Santa Fesina, Limeña, Torre blanca and Paulista.
Common names based on the person who made the intro auctions are Vargas, Valdivia, Pedrito, Remigio and Blank Salazar.
Names based on the similarity of the cultivar to other cultivated species, including traits like shape of the storage roots, skin colour, flesh colour and taste include remolacha; camote papa, camote yuca, batata abobora manicera, zapallo, batata zahanoria and camote oca.
Composite names generally refer to two traits of the cultivar such as blanco papa (white skin and flavour of Solanum potato), cargamontn dulce (good production an, sweet taste), morado harinoso (red skin and flour-like texture), colorada grande (red skin and large storage roots) roxa pintada (reddish skin and spotted flesh), morado mochero (reddish skin and flesh), amarillo rompe-costal (yellow flesh and good production), blanco matizado, (white flesh matrix with coloured spots), blanco grand (white skin and large storage roots), amarilla pequeña (yellow skin and small storage roots). Reasons for applying composite names to these plants probably vary al though most of these names distinguish between two similar local varieties.
Conclusions
The origins of the common names are related to characteristics of the cultivars, origin of movement, people who introduced the cultivar into a particular locality, and the strain's similarity to other cultivated plant species. Countries with the largest diversity of common names for the cultivars are Peru, with 394 different names, and Argentina with 115. Other surveyed countries have less than 100 common names in use for the sweetpotato.
In spite of the abundant variety of common names, comparisons of our findings with those presented in earlier studies (e.g. Yen 1974) show a marked decline in the number of names. We believe that this decline is linked with both the extirpation and extinction of local languages and the introduction of 'improved' lineages of sweetpotatoes. Both of these phenomena lead to biodiversity loss within the cultigen. Another difference between our data and those presented by Yen (1974) is that we did not target indigenous populations for census. However, we suspect that had less impact on our data than general biodiversity loss. The decline we found appears to be real and undoubtedly reflects the homogenization of human populations and cultigens that is in progress around the world. The numbers and types of geographical names in current usage support this interpretation.
Acknowledgements
The authors acknowledge financial support from IPGRI (Z. Huaman and F. De la Puente, principal investigators), CIP, USDA (D.F. Austin, principal investigator, Grant No. 586659-1-102) and the National Geographic Society (D.F. Austin, principal investigator, Grant No. 4478-91).
References
Austin, D.F. 1988. Taxonomy, evolution and genetic diversity of sweet potatoes and related wild species. P. 27-60 in Exploration, Maintenance and Utilization of Sweet Potato Genetic Resources, Proceedings of the First Sweet Potato Planning Conference, Centro Internacional de la Papa, Lima, Peru.
Brand, D.D. 1971. The sweet potato: an exercise in methodology. Pp. 343365 in Man Across the Sea (C.L. Riley et al., eds.). University of Texas Press, Austin.
Yen, D.E. 1974. The Sweet Potato and Oceania. Bishop Museum Press, Honolulu. pp. 1-389.
Résumé
Noms commune de la patate douce (Ipomoea batatas) dans les Amériques
Le présent article récapitule les différents noms commune donnés à la patate douce et à ses cultivars dans les pays d'Amérique latine et des Caraïbes. On a étudié l'origine de ces noms dans chaque pays et découvert qu'ils ont trait aux caractères physiques, aux populations, aux régions, aux localités et aux usages. En général, les noms commune vent déterminés par un seul critère.
Resumen
Nombres vulgares de la batata (Ipomoea batatas) en las Américas
Se hace un resumen de los distintos nombres vulgares dados a la batata y a sus cultivares en distintos países de América Latina y el Caribe. Se estudiaron los orígenes de esos nombres en cada país y se llegó a la conclusión de que guardan relación con las características físicas, la población, las regiones, las localidades y los usos. En general, los nombres vulgares están determinados en función de un único criterio.
Th. Gladis1, K. Hammer1, H.H. Dathe2 and H. Pellmann3
1 Institut fur Pflanzengenetik und Kulturpflanzenforschung Genebank, Corrensstr. 3, D-06466 Gatersleben, Germany
2 German institute of Entomology, Schicklerstr. 5, D-16225 Eberswalde, Germany
3 Department of Zoology, Leipzig University, Talstr. 33, D-04103 Leipzig, Germany
Summary
Cross-pollination in tomatoes in countries outside their natural distribution is a problem during ex situ seed regeneration in germplasm collections of wild and cultivated taxa. Isolation of different populations during the flowering period is one of the best methods to avoid hybridization. Bees and hover flies are efficient native pollinators of Lycopersicon species in Central Europe.
Introduction
Modern tomato varieties are pure breeds (monomorphic), predominantly self-pollinated, and there is assumed to be a very small amount of cross-pollination caused by insects. The outcrossing rate is correlated with the evolutionary age of the variety, i.e. with the position of the style within the flower. However, most wild tomato species (Lycopersicon spp.) and primitive types of cultivated or weedy plants, potato-leafed cultivars or forms with double blossoms, beefsteak tomatoes etc., are often long styled (Ashworth 1991). This paper reports observations on a large tomato collection maintained at the Institut für Pflanzengenetik und Kulturpflanzenforschung (IPK) genebank, Gatersleben, Germany. The implications for genebank work and breeding (isolation requirement) are discussed.
Multiplication of tomato collections under Central European conditions is considered a very easy task for cultivated material (Esquinas-Alcazar 1981). It only requires sowing or planting the accessions side by side, and harvesting them separately. The highest published crossing rate in Central Europe is 2%. Only in tropical areas may the cross-pollination rate attain 47% (Hackbarth 1960). Therefore, in northern countries the problem of foreign pollination is largely neglected and tomatoes are often treated as self-pollinating crops in breeding work and germplasm collections. Experience with other self-pollinating crops shows that there may be a large variation with respect to outcrossing tendencies, e.g. barley (Hammer 1975), and genebanks must develop specific methods to avoid unwanted intercrossings. Intensive entomological studies (Gladis 1989) and research on the floral biology of each plant population are preconditions for the work with entomophilous plants.
Observations
The IPK Gatersleben genebank maintains a living collection of more than 3050 tomato populations, i.e. varieties, mutants, unnamed populations, weeds and wild species. Each year about 300 accessions (average) are grown in reproduction fields and greenhouses for determination, multiplication and comparison of identity. A simple method is used to detect spontaneously occurring crosses, the so called disjunctive growing system as proposed by Lehmann and Mansfeld (1957): neighbouring populations carry specific morphological characters, sometimes making it easy to distinguish them and to detect outcrossings in the following generation.
Various characters in flower morphology were observed within one variety or population, within one plant, and even within one raceme (Fig. 1) regarding the length of the floral style. This has not yet been explained. Other observations after multiplication steps of genebank material led us to look for outcrosses after each seed-growing season, e.g. In 1976 a new variation within four older genebank accessions, out of 522 populations grown that year, was found and documented. Very often newly introduced material shows a high degree of heterozygosity when grown for the first time. This is attributed to occasional natural cross-pollination in southern countries (Lehmann and Schwanitz 1965). Lines derived from original material or self-pollinated species are maintained with higher priority than offspring of old varieties. They are always selected in the direction of the parental generation or characters of primary growth.
- Fig. 1. Various floral characters were observed within one tomato plant, and even within one raceme
Selection steps used to eliminate plants derived from outcrossings are always well documented, as the following examples illustrate: accession T 910/85 (first regeneration in 1995), in 1986 split into three different types (T 910/86, T 982/86, T 1025/86). In 1991 and 1992, two additional types derived from T 1025 (T 1206/91, T 1318/92). The original material (T 910) was collected in Italy in the Cosenza province in the region of Calabria in 1993. The indeterminate growing local variety with fruits of the San Marzano type led to additional morphotypes after ex situ reproduction steps at Gatersleben. This splitting seems to be more common in some years than in others. Comparable findings resulted from L. pimpinellifolium accessions, grown within the L. esculentum block. The following generations contained several intermediate plant types with different hair densities and fruit characters, which might originate from intercrossings: T 1127/88-93 segregated into T 1208, T 1209, T 1266, T 1279, T 1280 and T 1281). Characterization data, descriptions and herbarium sheets, altogether a complete reference file that includes photographs of the whole collection, may help with decisions about doubtful populations and duplications in many cases, but not in all. To study the responsible pollinators, field observations were carried out on insect densities in the tomato reproduction field in August 1993. That year was characterized by very inclement weather conditions for seed production as well as for flower-visiting insects. After a long period of low temperatures near the ground, 17-18 Aug were days with temperatures higher than 30°C at 5 cm above ground, the highest atmospheric pressure in the month (7.67 x 10-2 - 7.72 x 10-2 MPa), sunshine throughout the day, and less wind than other days. These two days were used to observe and to collect insects visiting tomato flowers.
Results
During the same period, the collecting behaviour of insects on tomato flowers was different from that concurrently observed on more attractive flowers (i.e. Apiaceae, Asteraceae, Brassicaceae, Campanulaceae, Fabaceae, Lamiaceae, Scrophulariaceae). Aggregations of hover flies, mainly Episyrphus balteatus, could be found peacefully feeding on these families. Solitary bees and honeybees collecting alone or jointly, often solitary bumblebees (seldom two or more workers on sunflower, Helianthus annuus L.), or females of larger Megachile species, being very aggressive to other insects on the same flower (erected abdomen with visual stinger) were also seen. The highest degree of intolerance is expressed by males of Anthidium manicatum. This common solitary bee mainly visits flowers of Fabaceae, Lamiaceae and some Scrophulariaceae. The males own and defend restricted areas where only females of the same species are allowed to expropriate the flowers of preferred plants. They also mate there. Competitors, whether capable of defending themselves or not, are turned out or killed. The predominance of different visitors on the same flower of, for example, Telekia speciosa (Asteraceae) or Verbascum sp. (Scrophulariaceae) followed more or less the same sequence (not seen in the case of Anthidium). In contrast to these observations, insects were never found fighting for Solanum or Lycopersicon flowers or those of related genera. The few insects collected on tomato flowers are listed in Table 1.
Table 1. Insects collected on tomato flowers (L. esculentum, L. pimpinellifolium), 17-18 Aug 1993
I. Hymenoptera, Apoidea: |
Months of activity |
Remarks |
Sol1 |
0:1_ Andrena bicolor (F.) |
Apr May, July-Aug |
Polylectic |
- |
0:l_ Apis mellifera L. |
Feb-Dec |
Polylectic |
+ |
0:3_ Bombuspascuorum (Scop.) Polylectic |
Mar.Nov |
Polylectic |
- |
0:1_ Bombus sylvarum (L.) |
Apr-Aug |
Polylectic |
- |
0:1_ Bombus terrestris (L.) |
Mar-Oct |
Polylectic |
+ |
II. Diptera, Syrphidae: |
|||
0:5_ Sphaerophoria scripta (L.) |
Mar-Oct |
Polylectic |
- |
_Sex ratio, male:female (1:2 means 1 male and 2 females)
_ 0,1 indicates 1 worker of a social species.
1 Sol +/reported as visitors on solanaceous flowers, yes/no
Discussion
Teppner (1993) observed Hylaeus gibbus as buzz pollinators of Lycopersicon flowers. Independent of the insect species, only females or workers, but never males, were found on tomato flowers. However, males of bumblebees were simultaneously present in other places. The females always have a greater need for pollen food than males, because of either their high egg production rate (hover fly) or their collecting larval food (bees).
All of the mentioned species are well-known as indefatigable flyers and pollinators. The observed behaviour of these insects on tomato flowers is quite different from their well-known behaviour on flowers of other plant families. Figs. 2 and 3 show how efficiently they collect tomato pollen. All insects stayed a very short time, had a swift and irresolute collecting behaviour and avoided distances far more than those for more attractive flowers. A comparable situation was seen in large fields planted for onion seed production (Allium cepa) near Quedlinburg, with Helophilus trivittatus (Syrphidae