University of the Orange Free State Department of Botany and Genetics, P O Box 339, Bloemfontein, 9300, South Africa.

Plant-aphid resistance.
Anna-Marie Botha-Oberholzer, Martie A.C. Nagel, and Alexander A. Myburg.
The aims of the programme are, first, the identification of molecular markers and the development of a key for cultivar identification and second, studying the mechanism of resistance in order to isolate relevant genes, which will be transferred into different susceptible wheat cultivars in order to confer resistance.
Identification of RAPD markers for Russian wheat aphid resistance in wheat. Six cultivars susceptible to and 13 cultivars resistant to the Russian wheat aphid (Diuraphis noxia) were used to identify both genotype specific, as well as D. noxia resistant, molecular markers. Fourteen polymorphic fragments, ranging from 300 to 1,200 bp, were identified from a total of 650 loci amplified with 60 decamer primers (Operon Technologies series A,B, and C). The linkage of these molecular markers to resistance genes, as well as their potential use in the wheat breeding programme, is being investigated currently.
Effect of exposure to the Russian wheat aphid on the expression of chitinase. Plants exhibit an altered pattern of protein synthesis in response to pathogen invasion and abiotic stress. One of these pathogenesis-related proteins has been identified as chitinase (EC 3.2.1.14.). Chitinase activity was determined in resistant and susceptible T. aestivum after Russian wheat aphid (D. noxia) infestation. Seven days after infestation by the aphid, a major induction in endochitinase activity was measured in the apoplastic fluid of the resistant lines. This induction in chitinase activity was not directly the result of wounding, because exogenously applied ethylene (15 nl/l) induced chitinase only in susceptible plants. Therefore, we speculate that infestation by D. noxia, ethylene, and mechanical wounding are separate, independent signals for the induction of chitinase.
Stem rust (Puccinia graminis f.sp. tritici).
J. Smith and B.D. van Niekerk.
During the 1994-1995 wheat production season, no stem rust infections were reported on commercial wheat fields in South Africa. Currently, the disease is under good genetic control. Although the annual rust survey for 1994-1995 season has not been completed, the results will be made available in the next Annual Wheat Newsletter. During the 1993-1994 season, two pathotypes with virulence to Sr27, i.e., 2SA102 and 2SA103, predominated, making up 86 % of the total population. The majority of races originated from the Southern Cape (43 %) and the Eastern Cape (29 %). During this season, a large increase in the incidence of stem rust was seen on trap nurseries in the Eastern Orange Free State because of favourable weather conditions during this season. Further stem rust research at the Small Grain Centre has the major objective of combining stem rust resistance genes with durable resistance. The main aim is to identify which genes have a complementary effect, and experiments on F₃ families will commence shortly.
Crater disease of wheat in South Africa.
E.M. Lubbe.
Crater disease (CD) occurs exclusively in dryland wheat cultivated on the black clay soils of the Springbok Flats, South Africa. It resembles bare-patch disease in respect to field symptoms, and in that the causal organism is Rhizoctonia solani. However, infection does not cause roots to rot, but characteristically produces nodulose swellings in roots, thereby inhibiting the uptake of water and nutrients. Survival of the pathogen is primarily by sclerotia. The CD pathogen does not anastomose with R. solani AG-8 and is related closest to AG-1, IA, and B. It has a limited host range within the Graminae, and appears to be indigenous to the Springbok Flats, but has not yet been isolated in this area from natural grass species in virgin soil. Nematodes are not involved in CD, but disease is more severe when wheat is co-inoculated with R. solani and a Pythium group F species. Effective control of CD was achieved by soil solarisation, treatment with phosphorous acid (H^3PO^3), and deep (65 cm) mouldboard ploughing. Rotation with nonhost crops for 2-4 years reduces disease appreciably, and is practised by farmers to control CD.
Entomology.
G.J. Prinsloo, V.L. Tolmay, and J.L. Hatting.
Further progress has been made towards the establishment of an integrated control programme against Russian wheat aphid.
Plant resistance to Diuraphis noxia has been incorporated into both winter and spring breeding programmes. Previously unused sources of resistance are being evaluted for efficacy and mechanism of resistance. A backcross breeding programme with five resistance sources and 10 cultivar parents is being undertaken. A further 40,000 Aphelinus hordei parasitoids have been released at four farms in the
eastern Free State. Parasitism levels of up to 53 % were recorded 2 weeks subsequent to release in these fields. During January 1995, A. hordei was recovered FONT SIZE=2 FACE="WP MathA"" 30 km from the place of release, which means that establishment had taken place. Therefore, the implementation of the integrated control programme against RWA has not commenced. This programme will reduce the RWA populations to extremely low levels, which will save the wheat farmers in South Africa more than R20 million per annum in chemical control.
Weed control research.
H.A. Smit, R.C. Lindeque, C.F. Pool, and B.L. de Villiers.
Various adjuvants and adjuvant combinations were tested to optimize glyphosate efficacy with a water carrier containing 20 mM calcium chloride. The combination of ammonium sulphate and methylated sunflower oil was the most effective adjuvant. Oils, in general, normally antagonize glyphosate efficacy. However, a possible mechanism for overcoming this antagonism in the presence of ammonium sulphate has not been postulated yet. Combinations of new and already registered herbicides were tested to minimize application costs. With the new adjuvants (methylated seed oil and ammonium nitrate) for tralkoxydim, two broadleaf herbicides (bromoxynil/MCPA ester formulation and an experimental formulation) could be used in combination. Tralkoxydim currently is registered in combination with the adjuvant AdditFONT SIZE=2 FACE="WP TypographicSymbols"7 and is phytotoxic on a range of wheat cultivars. In a study to optimize tralkoxydim efficacy, a most promising combination, namely tralkoxydim (150 g ai ha^-1), methylated seed oil, and ammonium nitrate, satisfactorily controlled wild oats without any phytotoxicity to wheat. The registered combination of chlorsulfuron/metsulfuron methyl and fenozaprop-P-ethyl was found to be antagonistic. The control of sweet buffalo grass with clodinafop propargyl was found to be effective. In a preliminary trial, we found that the residual action of the sulfonyl ureas reduces the effect of GauchoFONT SIZE=2 FACE="WP TypographicSymbols"7 on the R.
University of the Orange Free State
Department of Plant Pathology, Bloemfontein, South Africa.
Z.A. Pretorius and F.J. Kloppers.

Histological characterisation of resistance to wheat leaf rust.
In wheat genotypes resistant to Puccinia recondita f. sp. tritici, the absence of the hypersensitivity typically associated with Lr genes may be indicative of durability. To study whether leaf rust resistance genes transferred from alien sources to hexaploid wheat confer a hypersensitive reaction similar to those from other known sources, histological observations were made on Triticum tauschii, T. monococcum, and Aegilops speltoides. The wheat line KS90WGRC10 (TAM107*3/TA2460) (Lr41) and parent lines TA2460 (T. tauschii) and TAM107 (susceptible) were inoculated with South African pathotypes of P. recondita f. sp. tritici. Leaf sections were prepared for fluorescence microscopy. Histological observations showed that Lr41 conferred a high degree of resistance to leaf rust. Observations on Lr41 resistance predominantly noted early abortion associated with host cell necrosis. This post-haustorial resistance mechanism was similar to that observed in the diploid parent, TA2460, indicating that the mechanism remained the same in the hexaploid background.
Resistance transferred from Ae. speltoides was studied in germ plasm developed by G.F. Marais, Department of Genetics, University of Stellenbosch. This study included Ae. speltoides ssp. ligustica accession A1620, the T. durum cultivar Cando, the common wheat cultivar Pitic 62, and the hexaploid backcross derivative line F₄:91M56. Additionally, line RL6079 (Tc*/Lr28), which also derives its resistance from Ae. speltoides, was included. In A1620, a larger proportion of infection units and colonies aborted at an early stage of development compared to line F₄:91M56 where more colonies, with larger dimensions, developed. In both of these lines, a hypersensitive response was conspicuous. Resistance in F₄:91M56 was more effective than that conferred by Lr28 in RL6079, in which a large proportion of colonies developed into sporulating units.
Infection structures of P. recondita f. sp. tritici pathotype UVPrt9 in wheat line KS93U9 (Karl*3/PI266844/PI355520), developed by T.S. Cox at Kansas State University, were compared with those in the T. monococcum lines PI266844, PI355520, and Tm2126/5 and the susceptible cultivar Karl. The resistant lines responded similarly for prestomatal exclusion and abortive penetration, but differences in early abortion and colony formation were observed. All infection sites displaying early abortion in KS93U9 were associated with host cell necrosis, whereas this was not observed for Tm2126/5 seedlings. Staining of leaves with trypan blue and a saturated solution of picric acid in methyl salicylate showed that papillae commonly occurred at infection sites in the T. monococcum lines, but not in KS93U9.
Histological characterisation, especially hypersensitivity, suggested that Lr41 and the Ae. speltoides and T. monococcum-derived bread wheat lines evaluated, do not possess resistance mechanisms significantly different from those of other major Lr genes.

Germ plasm development. The first Palmiet and Karee selections, with either Lr29 or Lr34, were evaluated for leaf rust resistance in the field during 1994. Disease progress and terminal severities indicated that these lines contain high levels of leaf rust resistance. In an ongoing backcrossing program, Lr21, 22a, 32, 35, 36, 37, and 41 also are incorporated into the locally adapted cultivars Palmiet and Karee. All leaf rust-resistant lines resulting from this program are being evaluated for agronomic and quality characteristics by the Small Grain Centre in Bethlehem.
Publications.
Kloppers FJ. 1994. Characterisation of resistance conferred by selected Lr genes with emphasis on histopathology, leaf rust development and associated quality attributes in wheat. Ph.D thesis, University of the Orange Free State, Bloemfontein. 187 pp.
Kloppers FJ and Pretorius ZA. 1994. Expression and inheritance of leaf rust resistance gene Lr37 in wheat seedlings. Cereal Res Commun 22:91-97.
Pretorius ZA, Kloppers FJ, and Drijepondt SC. 1994. Effects of inoculum density and temperature on three components of leaf rust resistance controlled by Lr34 in wheat. Euphytica 74:91-96.
Pretorius ZA, Marais GF, Smith J, Le Roux J, and Kloppers FJ. 1994. Absence of the 1BL/1RS chromosome translocation in the Kavkaz-derived wheat cultivar Tugela. South African J Plt Soil 11:130-133.
SENSAKO, Welgevallen Experiment Station
University of Stellenbosch, 7600, Stellenbosch, South Africa.

R. de V. Pienaar and A.J.G. Lesch.
Doubled haploids (DH).
The production of DH from `wheat x maize' crosses was continued in 1994 using the best F<sub>4</sub> derivatives from the maize hybrid Seneca 60 as pollen parents. A total of 68,862 pollinated florets (PF) were injected with 70 wheat haploidizer (WH) solutions containing one or more growth regulators (GR). This resulted in 37,829 (= 54.9 % of PF) green parthenocarpic `caryopses' (GPC) 16-18 days after pollination; 13,685 (= 36.2 % of GPC) contained a haploid embryo (HE). By culturing these embryos on half-strength MS medium fortified with Gamborg's vitamins and 20 % sucrose (none of the other 40 media tested were significantly superior), 3,504 plants (= 25.6 % of HE or 5.1 % of PF) were obtained. Following colchicine treatment (see our protocol in AWN Vol. 40), some 90 % of these plants gave rise to DH.
The winter wheats, as a group, produced fewer haploid plants (4.1 % of PF) than the spring wheats (6.2 % of PF), but individual genotypes produced as much as 16 % haploid plants. Of the 30 spring wheat genotypes evaluated, only Pavon 76 and SST 55 produced >10 % (of PF) haploid plants during the best season. The F₂ from crosses between Triticum spelta and bread wheat cultivars yielded <1 % haploid plants. Haploid production by the chromosome elimination method is clearly influenced by the wheat genotype. Other factors of importance are season, maize genotype, and GR.
Using SST 55 to monitor the seasonal effect, the best results were obtained during the period from July-September (winter to early spring in the southern hemisphere), i.e., 86 % GPC, 40 % HE, and 11 % plantlets (as a percent of PF using the best WH solution). The results of the period from March-May were 73 % GPC, 26 % HE, and 8.5 % plantlets; similar results were obtained during the period of October-December. The other wheat genotypes produced fewer GPC, HE, and plantlets during the summer season. The HE frequencies obtained by using 20 individual maize plants as pollen parents, varied from 12 % to 33 % (of PF). Therefore, it is advisable to use pollen mixtures from at least two or three maize plants in haploid induction experiments. Pollination with maize on 2 consecutive days did not increase the number of HE compared to those obtained with a single pollination.
The 70 haploid inducing aqueous solutions tested in 1994 contained higher concentrations of GR than those tested previously. This was aimed at overcoming the time-consuming and hazardous injection of the solution in the internode below the spike. GR such as FONT SIZE=2 FACE="WP Greek Century""-NAA, PAA, picloram, and 2,4,5-T, not tested previously, were included in various solutions. The WH solutions that gave the best results (when only the floral cups were injected 24-30 hours after pollination with maize) for the period of March-December on both spring and winter wheats are as follows:
1. WH5 = 30 mg/L dicambra,
2. WH6 = 30 mg/L dicambra + 10 mg/L BA,
3. WH7 = 30 mg/L picloram,
4. WH8 = 18 mg/L dicambra + 12 mg/L picloram, and
5. WH9 = 10 mg/L dicambra + 20 mg/L 2,4-D + 3 mg/L BA.
These solutions produced results equal to, or better than, that of WH1-WH4, which had to be injected into the floral cups, as well as the internode below the spike (see AWN Vol. 40). WH8 gave the most consistent results during winter and spring. The solutions containing FONT SIZE=2 FACE="WP Greek Century""-NAA and PAA did not induce any GPC, whereas 2,4,5-T produced satisfactory results only on SST 55.
It is evident that sufficient HE can now be produced to make DH breeding worthwhile. The major remaining problem is the low percent (FONT SIZE=2 FACE="WP MathA"" 25 %) of HE that differentiate into plantlets. Future research should try to overcome this problem. Medium composition has little effect; most of the media used gave similar results. An improvement in HE `germination' was obtained when the cultures were kept in the dark at 4-6 C for 6-8 weeks.
University of Stellenbosch
Department of Genetics, Stellenbosch, 7600, South Africa.
G.F. Marais, R. Prins, A. Antonov, H.S. Roux, and A.S. Marais.
Durum wheat breeding.
Very little durum wheat was produced locally for the past 4 years, because producers could not compete with subsidised durum and durum products imported from EEC countries. As a result, we had to curtail the breeding programme to a maintenance level. The situation is expected to gradually return to normal over the next 2-3 years, and the programme is being scaled up to its former proportions. We select fast maturing, high quality spring genotypes for intended production in the lower Orange River irrigation areas.
Triticale breeding.
The programme aims to develop spring triticales for dryland cultivation in the western, southern, and eastern Cape Province. The cultivar `Rex' was released recently for dual purpose cultivation and has been received very favourably by farmers. It outyields the existing cultivars by about 10 %. Triticale grain is being used increasingly to substitute for some of the maize in animal rations, and, as a result, the acreage sown to triticale continued to expand in 1994. It is estimated that the demand for triticale grain will grow by about 30 % annually to reach about 75,000 tonnes by 1997. Triticale cultivation techniques also have improved significantly, and the crop can be grown very successfully on the marginal soils.
Cytogenetics.
A wild species collection was screened for resistance to an inoculum mix of five leaf rust races. A total of 240 accessions (232 from the genera Triticum and Aegilops, and 8 from Thinopyrum) proved to be resistant to all races. One hundred and nine resistant accessions have not yet been crossed successfully, whereas the remaining 131 (representative of 20 species) were crossed with common wheat, durum wheat, or 'Tetra' Chinese Spring, depending on the ploidy level of the wild parent. The interspecies hybrids had mostly distinct phenotypes or were validated by doing root tip chromosome counts. A number of transfer attempts failed in the F₁ as a result of suppression or irregular expression of the resistance (54 accessions of T monococcum, T turgidum, T timopheevii, T syriacum, T triunciale, T ovatum, and T longissimum); formation of embryoless seeds (10 accessions); or poor F₁ viability (10 accessions). In 44 hybrids, the resistance is fully expressed, and these are now in various stages of backcrossing to wheat. In some instances, the chromosome numbers of the hybrids had to be doubled beforehand to ensure fertility during backcrossing. Two of the sources have been backcrossed to wheat once, 29 twice, 10 three times, and 3 were backcrossed four times. Backcrossing will be continued until a stable hexaploid or near hexaploid wheat background has been restored.
The stem rust resistance gene, Sr27, was transferred from chromosome 3R of Imperial rye to chromosome 3A of common wheat in 1962 (Acosta). However, this irradiation-induced translocation (WRT238) reduced yield in its homozygous state and was never used commercially. We found that the detrimental effect of the translocation results largely from it being noncompensating, combining chromosome arm 3RS of rye with 3AS of wheat. Therefore, an attempt was made to purposely derive compensating wheat-rye translocations having Sr27. Making use of the appropriate Chinese Spring monosomics, and a line having a disomic addition of Imperial rye chromosome 3R, F₁ plants that were double monosomic for chromosome 3R and either chromosome 3A or 3B of wheat were obtained. These plants were either left untreated or irradiated (7.5 Gy gamma rays) before being used to pollinate susceptible testers. Based on the segregation of Sr27 in the testcross progenies, the presence of translocations could be detected, and they then were identified through C-banding. Starting with 219 resistant testcross F₁ families, 19 compensating (3RS-3AL and 3RS-3BL) and 3 noncompensating (3RS-3AS and 3RS-3BS) translocations were found. In 82 % of the cases, translocations occurred between the monosomic chromosomes; in the remaining instances, a wheat disomic chromosome was involved. Irradiation appeared to be unnecessary, because it had no apparent effect on the frequency with which compensating translocations were obtained. Through in situ hybridization with rye genomic DNA, Dr B. Friebe (KSU, Manhattan, KS, USA) demonstrated that 20 of 21 of these translocations were Robertsonian.
A study of the Lr19 translocation in the germline `Indis' was continued. Making use of further marker genes, a series of deletion mutants and ph-induced recombinants was studied. The data suggest that the Lr19 resistance may be determined by two genes rather than one. Also, segregation distortion apparently results from the action of at least three, and possibly four, Sd loci. Loss of one or more of these loci may sometimes result in self-elimination depending on the nature of responder genes present in the wheat genomes. The specific combination of Sd genes retained in one of the white endosperm recombinants makes this shortened segment particularly prone to self-elimination. In a study of two B<sub>6</sub>F<sub>2</sub> populations (recurrent parents `SST 66' and `W84-17'), the following genotypic ratios were found: 1 RR : 53 Rr : 46 rr (125 plants were tested). An attempt is being made to derive lines each carrying the Lr19 gene singly.
Genes for resistance to leaf and stem rust derived from a T. speltoides accession appear to be associated with a gametocidal locus having a very severe expression. Gametes without the Gc gene do not survive, and the Gc-associated defects (low fertility, poor seed development, chromosome instability, and a high mutation rate), therefore, are always inherited with the resistance. Following irradiation and treatment with N-nitroso-N-methyl-urea, a very low frequency of the treated material showed slightly improved fertility and seed quality. However, these effects mostly did not persist in subsequent generations. It is possible that the Sd system may be multigenic and/or that it is situated proximally to the resistance loci, whereas the mutagen treatments employed may have caused mostly terminal deletions. Renewed attempts will be made to disable the Gc system using the better selections as starting material.
In a telocentric study of the Russian wheat aphid resistance gene, Dn5, we found that the resistance is associated with chromosome arm 7DL of wheat. Possible linkage with the Cn-D1, Ep-D1, or Lr34 loci is being investigated.
Publications.
Marais GF and Marais AS. 1994. The derivation of compensating translocations involving homoeologous group 3 chromosomes of wheat and rye. Euphytica 79:75-80.
Marais GF, Horn M, and du Toit F. 1994. Intergeneric transfer (rye to wheat) of a gene(s) for Russian wheat aphid resistance. Plant Breed 113:265-271.