African Crop Science Journal African Crop Science Society ISSN: 1021-9730 EISSN: 2072-6589 Vol. 5, Num. 1, 1997, pp. 9-14

Evaluation of accessions of South African sorghum germplasm for use in the development of improved varieties

W.G. WENZEL, J. MOHAMMED^1 and J. VAN DEN BERG

ARC- Grain Crops Institute, Private Bag X1251, Potchefstroom 2520, South Africa
^1 Crop Science Department, National University of Lesotho, Roma, Lesotho

(Received 26 March, 1996; Accepted 12 February, 1997)

Code Number: CS97002
Sizes of Files:
    Text: 15.1K
    Graphics: Tables (gif) - 30.4K

In order to identify better adapted open-pollinated sorghum (Sorghum bicolor L. Moench) varieties for small-scale farmers in Southern Africa, 1825 accessions of the South African germplasm collection and other promising varieties were screened in South Africa and Lesotho. Promising genetic sources for ear size and grain quality characteristics, such as hardness, colour and size, as well as resistance to aphids, stem borers and grain moulds were identified. It was concluded that the screening and further characterisation of the South African collection for use by breeders could be a contribution to curb the declining sorghum production in the region.

Key Words: Early maturity, aphids, grain quality, landraces

RESUME

Pour mieux identifier les varietes de sorgho ouvertement-pollinisees et bien adapte pour les petits agriculteurs en Afrique australe, 1825 accessions de la collection de germoplasm de l'Afrique du Sud et autres varietes prometteuses etaient evalues sur le champs en Afrique du Sud et au Lesotho. Des sources genetiques prometteuses en ce qui concerne la taille de l'epi et les caracteristiques qualitatives des graines, comme la durete, couleur, et taille, mais aussi la resistance contre les pucerons, les foreuses de tige et moisissures des graines etaient identifiees. On a conclu que l'evaluation et la caracterisation de la collection Sud Africaine pour l'utilisation dans l'amelioration peuvent contribuer ˆ controler la reduction de la production de sorgho dans cette region.

Mots Cles: maturite precoce, puceron, qualite des graines, varietes

INTRODUCTION

Sorghum (Sorghum bicolor L. Moench) is an important grain crop in the semi-arid tropics. It is usually grown in areas too dry and hot for maize production. In South Africa, sorghum is grown by commercial and small-scale farmers. Commercial farmers grow hybrids only and the largest percentage of the harvest is used for the production of opaque beer. Sorghum varieties are grown by majority of the small-scale farmers in Botswana and Zimbabwe (Manthe, 1992). In South Africa, traditional land races of sorghum are grown mostly as a mixture with or without maize and with cowpeas and watermelons. The maize component is reduced or absent in drought years (Van der Maesen, unpubl.). Most of the South African small-scale farmers have changed to maize production mainly due to bird depredation. However, sorghum is still an important cereal for small-scale farmers due to its drought resistance and uses in traditional food products.

The sorghum breeding programme at the Grain Crops Institute is aimed at developing improved inbreds and hybrids for commercial farmers and varieties for small-scale farmers. It is anticipated that all farmers will eventually plant hybrids due to their (hybrids) superior adaptation and yield. The local varieties were selected over centuries for their improved food quality and other kernel characteristics such as size, grain mould resistance and high flour extraction. The genes that transmit superior kernel characteristics will be used to develop new, improved varieties and will also be introgressed into new inbred lines for the development of improved hybrids.

The objectives of the small-scale farmers sorghum breeding programme are (1) collection of a wide range of germplasm, (2) identification of superior plant and seed characteristics, and (3) development of improved varieties and inbred lines that are adapted to low-input farming systems, posses the desired food quality and are tolerant to biotic and abiotic production constraints. This report describes this programme which was initiated in 1992.

MATERIALS AND METHODS

Germplasm collection. For a long time, farmers have continued to select sorghum based on desirability for human consumption (Mann et al., 1983). In the development of the South African germplasm collection initiated during 1983, the objectives were twofold. Firstly, it was necessary to preserve the unique land races developed by traditional farmers. Secondly, by preserving these land races, we intended to preserve the genetic diversity found within these accessions so that they could be utilised in the development of new, improved varieties, inbred lines, or hybrids.

Accessions to the germplasm collection were obtained from various sources (Wenzel, 1995). Large contributions were made by the Botanical Institute (Arnold and Musil, 1983), the International Crop Research Institute for the Semi-Arid Tropics from their South African collection (Van der Maesen, unpubl.) and from other collections or improvement programmes and various countries (Wenzel, 1995).

Identification of superior landraces. During the season, from 1992/3 to 1994/5, the following accessions were planted in single rows for preliminary evaluation at Potchefstroom, Maseru, and Siloe under rain-fed conditions. In the 1992/3, 1332 accessions from local germplasm collection were grown at Potchefstroom only but in 1993/4, 77 accessions selected during the previous season, 54 accessions developed locally and 105 accessions from South Africa were screened at Maseru and Siloe. In the 1994/5 season, 131 accessions selected during the previous season, 141 accessions obtained from the National University of Lesotho, and 193 varieties from South Africa were screened at Maseru and Siloe.

Potchefstroom in South Africa has a well maintained soil and relatively dependable rainfall, whereas Siloe and Maseru in Lesotho have eroded soil (low in organic matter) and are prone to drought conditions. The length of the growing period at the two locations is determined by the onset of rain in October/November and the onset of low temperatures in April. Hence, early maturing genotypes are preferred (Mohammed, J., unpubl.).

The varieties were screened during each season for earliness, plant height and head size, but seed size, seed colour and plant colour were determined during the 1994/5 season only. Five seeds of each variety were cut longitudinally and the hardness or texture of the endosperm was assessed on a scale of 1 (100% glassy) to 5 (100% floury).

Seeds of the most promising varieties selected for their agronomic suitability, that is, plants with large panicles, short stature (<2.0 m), no lodging, tan plant colour, were analysed for diastatic power according to the procedures described by Novellie (1962). Seeds (200 g) were steeped for 20 hr followed by malting for six days and drying of the malt. The diastatic power was determined on peptone extracts of the dry malt.

Grain mould resistance was evaluated during the 1994/5 season only. Frequent rain during the seed development period provided excellent conditions for grain mould development and identifying those varieties exhibiting resistance. Seed samples were visually evaluated. The varieties that produced seeds free from any discolouration were considered possible sources of resistance.

The site where the accessions were planted is considered a Òhot spotÓ for natural infestation of the honey dew aphid, Melanaphis sacchari. Resistance was evaluated on a scale of 1 (no infestation) to 5 (severely infested). Those with a mean rating of < 1.5 were classified as resistant.

In order to screen for stem borer resistance, five plants of each plot were artificially infested with larvae of Busseola fusca during the 1992/3 and 1994/5 seasons. A leaf rating of < 5 on a scale of 1 to 9 was indicative of potential antibiosis resistance (Van den Berg et al., 1994). Those that produced panicles normally were considered tolerant.

Development of improved genotypes. The pedigree and backcross selection methods were used to develop new, improved varieties. Crosses were made between locally adapted inbred lines and varieties and sources for cold tolerance, malt quality, flour extraction, ergot resistance, stem borer and aphid resistance, tan plant colour, short plant stature, grain yield and earliness. Progenies were planted in the field and agronomically acceptable F2 plants were selected.

RESULTS AND DISCUSSION

The South African germplasm collection presently consists of 4000 accessions of which 1244 were collected locally (Wenzel, 1995). Most are land races grown by farmers who selected them for specific attributes over the years. These land races are, therefore, an important source for good quality and resistance to stress, such as diseases, insects and drought. This is in part illustrated by the malt quality result obtained during the 1992-1994 seasons (Table 1). The mean sorghum diastatic units (SDU) of the varieties (40.0) was higher than those of the B-lines (32.7) and R-lines (22.6). The varieties are used as sources of malt quality for the development of improved varieties and inbred lines. The malt quality of a genotype is largely influenced by the environment.

Table 1

Sorghum genotypes have also been imported from many countries for specific attributes such as resistance to disease, insects, Striga, drought and birds, acid soil tolerance, low hydrocyanic acid content, high flour extraction, cold tolerance and grain yield (Wenzel, 1995). Germplasm exchange programmes exist and are being extended. This is in part exemplified by the multi-national contributions of the selected accessions presented in Table 2. New varieties and inbred lines are continuously being developed. The germplasm collection presently constitutes a vast source of genetic variation from which selections can be made for specific objectives.

Over the three seasons, a total of 1825 varieties were screened for agronomic acceptability and other characteristics. From these trials, 23 and 14 elite lines (Table 1 and Table 2, respectively) were selected which were viewed as very promising and others (Table 3) that were promising enough to merit further evaluation. The evaluation programme will have three phases, namely, (1) the preliminary and local evaluation of new entries which have been collected, imported or developed in the breeding programme; (2) evaluation on a national basis at different localities; and (3) evaluation on a regional basis, involving different countries in the Southern African Region.

Table 2

Table 3

Sorghum varieties from the South African germplasm collection have performed well in Lesotho. Out of the 105 lines screened at Maseru during the 1993/4 season, six lines were retained because they were less susceptible to birds. An additional 16 genotypes were retained for good agronomic traits at the same location. In the more moisture stressed areas of Lesotho, 11 genotypes were retained at Siloe for early maturity and good grain quality. The good performance of most varieties from South Africa has suggested that emphasis should be given to the introduction and screening of more materials from this source (Mohammed, J., unpubl.).

An additional 193 genotypes from South Africa were screened at Maseru and Siloe during the 1994/5 season. Results have revealed that sorghum genotypes from South Africa performed well and appeared promising in Lesotho. Over 50 genotypes were selected and retained for further studies. Growing conditions were particularly unfavourable at Siloe and the selected materials did well despite late planting (Mohammed, J., unpubl.).

The results given in Tables 1, 2 and 3 indicate the existence of promising sources for seed size, grain mould resistance, malt quality, stem borer and aphid resistance. Further evaluation will be required to verify these results.

The decline in the cultivation of sorghum in Southern Africa is mainly attributed to crop failure due to drought and bird depredation. Resistance to birds and early maturity are, therefore, of cardinal importance. Several accessions of the South African germplasm collection matured in about hundred days and were much less attractive to birds. Several of these promising accessions are currently being incorporated into the breeding programme with the aim of developing new, improved germplasm. The screening and further characterisation of the South African collection for use by breeders will contribute towards curbing the declining sorghum production in the region.

ACKNOWLEDGEMENT

We are grateful to all who contributed towards the germplasm collection. Malt quality tests were carried out by Mrs E. Stander and Miss G.J. Muller. The assistance of J. Cilliers, J. Kolodi and J. Jafta is gratefully acknowledged. Insects were reared by Mrs U.M. du Plessis.

REFERENCES

Arnold, T.H. and Musil, K.J. 1983. A preliminary survey of primitive crops cultivated in the northern Transvaal of South Africa. Bothalia 14:595-601.

Mann, J.A., Kimber, C.T. and Miller, F.R. 1983. The origin and early cultivation of sorghums in Africa. Bulletin 1454, Texas Agric. Exp. Station.

Manthe, C.S. 1992. Sorghum Resistance to the Sugarcane Aphid (Homoptera: Aphididae) Ph.D. Thesis, Texas A&M University.

Novellie, L. 1962. Kaffircorn malting and brewing studies. XI. Effect of malting conditions on the diastatic power of kaffircorn malts. Journal of Science and Food Agriculture 13:115-120.

Van den Berg, J., Wenzel, W.G. and Van der Westhuizen, M.C. 1994. Tolerance and recovery resistance of grain sorghum genotypes artificially infested with Busseola fusca (Fuller) (Lepidoptera: Noctuidae). Insect Science and its Application 15:61-65.

Wenzel, W.G. 1995. The South African germplasm collection. International Soghum and Millets Newsletter 36:12-13.

Copyright 1997 The African Crop Science Society

The following images related to this document are available:

Line drawing images