ABSTRACT

Ethiopian mustard (Brassica carinata Braun) occupies a significant place in the farming systems of Ethiopian agriculture. For the last two decades, efforts have been made to develop cultivars with better agronomic and nutritional attributes. All the cultivars developed thusfar, however, are shunned because of high erucic acid in their oil and glucosinolate in their meal. In order to address the ever-increasing demand for high-quality cultivars, assessments have been made with the germplasm material in Göttingen, Germany. Erucic acid ranged from 28 to 41% in the collections and from 24 to 46 in the successively selfed lines. The glucosinolate content varied from 53 to 158 μ mol/g of seed in the former and from 80 to 165 in the latter. If there exist different alleles controlling lower values and if their subsequent combinations will result in further lower values is a subject under investigation.

INTRODUCTION

The culture and cultivation of Ethiopian mustard (Brassica carinata Braun) in Ethiopia is as old (if not older) a tradition as cultivation of cereals, which is believed to date back in the 4^th to 5^th Millennia BC. Traditional utilization of this crop embraces quite an array of purposes. Ground seeds are usued to grease a bread-baking clay pan, cure certain ailments or stomach upsets and prepare beverages; the leaves of young plants are good source of vegetable relish. The oil, very often adulterated with the premium oil from niger seed is the commercial product. One very important additional advantage in the farming systems, especially in respect of the growing large-scale farms, is the role it can play as a break crop for the cultivation of cereals with comparable ecological amplitude.

Despite its long history and deep-rooted tradition of production, however, until very recently it has never been known as a full-fledged field crop. Its cultivation was so limited that it was grown either as a garden crop around homestead or sparsely mixed within thick crop stands of maize, sorghum, tef and finger millet. Since about a decade and a half, it is gaining a momentum of being grown in larger farm lands of pure stands. The reasons could be of both social and economic grounds, but the fact remains that there is today a steady increase in both production hectarage and tonnage.

Conversely, however, the cultivars under production are shunned because of their low-quality oil and meal which correspondingly are attributed to high levels of erucic acid and glucosinolates. Despite the fact that there is a tremendous amount of germplasm material collected from different corners of the country, the technical information on the extent of its genetic diversity and the opportunity thereof to exploit the genepool inline with improving these two quality characteristics is virtually absent. This particular experiment was, therefore, conceived to generate information that could be used to mitigate the gap between the ever-increasing demand for high-quality cultivars in the production sector in one hand and lack of developing such cultivars in the research arena on the other.

MATERIALS AND METHODS

Two hundred and thirty-six accessions of Ethiopian mustard (Brassica carinata) were taken to the Institute of Agronomy and Plant Breeding of the University of Göttingen, Germany. These materials were all obtained from the National Collections of the Ethiopian Biodiversity Institute. They are believed to represent the major regions of the genetic resource in the country. The accessions were all subjected to the analyses of fatty acids, glucosinolates and oil and protein contents using near infrared spectroscopy (NIRS) (Velasco, et al 1996). In order to further confirm the NIRS' values, 57 selective accessions were analysed by gas chromatography (GC) for fatty acids and by high-pressure liquid chromatography (HPLC) for glucosinolates. These accessions were selected based on the range of each trait so as to capture the whole range of values of high, low and intermediate categories.

Within each category, inbred lines have been developed by continuously selfing and selecting individual plants over several generations. Selections were made based on the analyses of half-seed technique for fatty acids (Thies, 1971), HPLC for glucosinolate and NIRS for oil and protein contents mainly based on their sum as is suggested by Velasco, et al (1997).

RESULTS AND DISCUSSION

It could be discerned from Tables 1–3 that there is a considerable amount of variations in fatty acids, glucosinolate, oil and protein contents both in the germplasm as well as in the inbred lines. The level of erucic acid is particularly so wide ranging from 24 to 46% in the inbred lines and from 28 to 41% in the germplasm material (Table 2). It is even more interesting to note that the average erucic level of the selectivity analysed accessions and lines which is about the same 37% for both sources. This is in fact much lower than the level typical for carinata (Velasco, et al 1995) implying the fact that genotypes with low erucic acid content are not uncommon in the genepool of Ethiopian material. It has been suggested by Rakow (1995) with the mutant plants of 13% erucic acid developed in Cordoba, Spain that zero-erucic recombinants could be recovered. By the same talken there could be a possibility of identifying genotypes with further lower erucic acid content by intercrossing the inbred lines.

The values obtained with the analyses of glucosinolates in the germplasm seem a little bit far from realistic which could be attributed to environmental factors (Table 3). The results with the inbred lines, on the other hand, are very much like the mutants reported from Cordoba and reviewed by Rakow (1995). In both cases, these results reveal the diversity of the genepool which could be of significant importance in respect of developing low-glucosinolate genotypes. Same line of argument and prospect holds true for both oil and protein contents, too.

Table 1. Variabilities in fatty acid compositions of germplasm (GPLSM) accessions and inbred lines (INBL) of Brassica carinata as were determined by near infra-red spectroscopy (NIRS).

Table 2. Variabilities in fatty acid compositions of germplasm (GPLSM) accessions and inbred lines (INBL) of Brassica carinata as were determined by gas chromatography (GC).

Table 3. Variabilities in glucosinolate, oil and protein contents of germplasm (GPLSM) accessions and inbred lines (INBL) of Brassica carinata as were determined by near infra-red spectroscopy (NIRS) and/or high-pressure liquid chromatography (HPLC).

ACKNOWLEDGEMENTS

The authors are grateful to the German Academic Exchange Service (DAAD), Institute of Agronomy and Plant Breeding of the University of Göttingen and the Ethiopian Agricultural Research Organization. Special thanks are due to Dr Leonardo Velasco for his unreserved help with the analyses by NIRS.

REFERENCES

1. Rakow, G. 1995. Developments in the breeding of edible oil in other Brassica specie. Proceedings of the 9^th International Rapeseed Congress, D8.

2. Thies, W. 1971. Schnelle und einfache Analysen der Fettsäurezusammensetzung in einzelnen Raps-Kotyledonen I. Gaschromatographische und papier chromatograpphische Methoden. Zeitschrift für Pflanzenzüchtung 65:181–202.

3. Velasco, L., Fernandez-Martinez, J.M., DeHaro, A. and Del Rio, M. 1995. Obtainment of EMS-induced mutants in Ethiopian mustard (Brassica carinata Braun). Proceedings of the 9^th International Rapeseed Congress, D14.

4. Velasco, L., Fernandez-Martinez, J., Haro, A-de and DeHaro, A. 1996. Screening Ethiopian mustard for erucic acid by near infrared reflectance spectroscopy. Crop Science 36(4):1068–1071.

5. Velasco, L., Fernandez-Martinez, J., Haro, A-de and DeHaro, A. 1997. Selection for high sum of oil and protein in Ethiopian mustard (Brassica carinata Braun). Cruciferae-Newsletter No. 19:97–98.