INTRODUCTION

The regeneration of potato plants from somatic tissues such as leaf mesophyll protoplasts is now well established (Gunn 1982). Examination of the somaclones raised from these tissues has revealed variation for a range of characters, some such as resistance to common scab (Streptomyces spp.) being of economic importance in potato breeding (Gunn et al. 1985). However it has been demonstrated that the variation occurs only for a limited number of characters and in many respects the somaclones are identical to the parental variety from which they were derived (Secor & Shepard 1981). This suggests that it may be possible to upgrade existing varieties by selecting amongst somaclones for improvements in those characters in which the parental variety performs poorly while maintaining the same level of character expression for those characters in which the parental variety performs well (Thomson et al. 1986). Some recent data obtained to explore this possibility are reported here.

MATERIALS AND METHODS

The data were recorded from field trials of somaclones derived from either Feltwell or Maris Piper which are varieties bred at the Plant Breeding Institute, Cambridge. The trials comprised either two or three replicates of six-tuber plots grown in 1984 and 1985. Total harvested yield and general tuber appearance were recorded. In 1985 the produce from those somaclones selected from the 1984 trials was riddled to obtain the different size grades of the tubers.

Potatoes are now beginning to receive special attention from molecular biologists and tissue culture experts who are committed to introducing new genes from the test tube and establishing the principles of gene regulation. This is because the potato is one of the few crop species that is infected by Agrobacterium tumefaciens, the organism which can transfer new genes into chromosomes of many dicotyledonous species; plants can also be regenerated from the single cells that have received the new genes, although this latter property is still in need of considerable improvement for routine genetic engineering of the crop. The status of current research is covered by other contributions to this volume, in particular the papers by Ooms and Blau et al.

Potato breeders can be excited that their economically important crop has become the experimental organism of a new group of scientists eager to try out new techniques and ask fundamental questions on the frontiers of academic research. It is inescapable that potato breeding will be influenced by this research.

The introduction of new genes into potatoes is in its infancy but progress in learning how to do it will surely be rapid over the next few years because some of the knowledge (and many of the genes) being gained from studies on tobacco is transferable to potato.

INTRODUCTION

Temperature pretreatment of excised floral parts is a technique used to improve response to anther culture in many species. In some solanaceous species including Solanum tuberosum (Uhrig 1983) pretreatment has involved placing the excised buds at low temperatures, usually around 6°C, for 2–7 days prior to culture. However, more recent investigations on Capsicum annuum (Dumas de Vaulx et al. 1981) have shown that for this species, high temperature pretreatment during the first days of culture (i.e. preincubation) gives optimum results. The first report of preincubation at high temperatures (30°C, 2 days) for a tuberous Solanum species was by Cappodocia et al. (1984) on S. chacoense and various hybrids. However, no comparison was made between this treatment and the cold pretreatment described above. The present study involved the dihaploid S. tuberosum H37O3 and compared the effects of various pretreatments before and during culture and included the use of both high and low temperatures in two experiments.

MATERIALS AND METHODS

Scions of S. tuberosum cv H37O3 were grafted onto tomato rootstocks to promote and prolong flowering. For the first experiment plants were grown in growth cabinets at 20°C day, 16°C night, 18h day. Illumination was provided by fluorescent tubes and 25W tungsten filament lamps (5000–8000 lux). For the second experiment the same procedure was adopted except that plants were glasshouse grown.

INTRODUCTION

Potato breeding involves selection over several years from a large number of clones, and the breeder has many varieties to maintain. Seed numbers are inevitably low for both practical and economic reasons. This is often a limiting factor in the progress of a variety through advisory and developmental trials. It is therefore advantageous to accelerate seed production at this stage by means of micropropagation.

The Potato Section of the National Institute of Agricultural Botany (NIAB) produces its own seed of new varieties for Recommended List trials and other advisory trials. Prior to 1982 primary propagation was carried out by virus tested stem cuttings (VTSC); this gave improved multiplication rates over tubers and the plants were free from pathogenic fungi (Hirst & Hide 1966). In 1982 NIAB replaced stem cuttings with in vitro propagation (micropropagation); this gave better control of multiplication, improved hygiene and easier transportation of material. Micropropagation reduced seed production time for most varieties by one year.

METHOD

The method consists of a phase of micropropagation followed by transfer to soil under glass and finally transplantation into the field (Wooster & Dixon 1985). Polythene tunnels or screenhouses may be used when field conditions are not suitable for high value seed.

Laboratory stage

The potato is relatively easy to propagate in vitro using axillary buds as the basic propagule. Initial explants can be taken from plant stems or tuber sprouts.

HISTORY

The potato was introduced into Europe from South America sometime between 1565 and 1573. It was first grown in Spain but by the early part of the seventeenth century it was found in the botanical gardens of many European states. This was largely due to the efforts of the botanist Charles d'Ecluse, or Clusius, who received two tubers and a fruit from the Prefect of Mons, Philippe de Sivry in 1588, multiplied the tubers and distributed them to a number of friends. The first botanical description of the potato was published by the Swiss botanist, Caspar Bauhin in 1596. Bauhin also gave the potato its Latin binomial, Solanum tuberosum, although he later added esculentum. A second description was published in England, where the potato was probably introduced between 1588 and 1593, by Gerard in his 1597 “Herball”.

The original introductions were almost certainly from the Andean regions of Peru or Columbia and were of the subspecies andigena (hereafter called Andigena potatoes), with a short-day photoperiodic response. Under European conditions they would have produced many stolons but poor yields of late-developing tubers. During the following two centuries, seedlings from these original introductions were selected for yield and earliness, to give rise to clones well adapted to longer summer days. This selection may have been made in hybrids and selfs of only two original introductions (Salaman 1926).

Research at the SVP on resistance to Phytophthora infestans focuses on:

1. 1) the development of efficient methods to evaluate the level of race nonspecific haulm resistance to P. infestans and to distinguish between different components of resistance;

2. 2) the detection of genotypes with race nonspecific resistance to P. infestans;

3. 3) the study of resistance mechanisms and the inheritance of the resistance; In this paper these research topics are highlighted and future research plans are mentioned.

DEVELOPMENT OF TEST METHODS

A field test has been developed at the SVP to assess leaf resistance to P. infestans. Clones to be tested are planted between spreader rows of a susceptible variety, in replicated microplots of six plants each. Standard varieties are included for comparison. All plants are inoculated artificially with a zoospore suspension. The percentage of diseased leaf area is assessed seven times per clone at weekly intervals, starting one week after inoculation. These seven scores are transformed to a weighted mean (WM) value using a formula (Tazelaar 1981). The WM values of 30 varieties lacking major (R) genes corresponded well with the known scores for resistance to P. infestans in the Dutch variety list (r = -0.85). Moreover the results of this test appeared to be highly reproducible.

However, correlation analyses showed that there are high positive coefficients of correlation between:

INTRODUCTION

It is well known that haploids extracted from Solanum tuberosum ssp. tuberosum provide unique opportunities for germplasm transfer and genetic manipulation.

Haploid extraction in potato has become a routine method since Hougas & Peloquin (1957) showed that haploids are relatively easy to obtain from 4x x 2x crosses.

Haploids can be easily hybridized with most 24-chromosome tuber bearing species. The hybrids obtained are vigorous, fertile and have an improved tuberization under long-day conditions. Hermundstad (1984) found that many of the haploid tuberosum-species hybrids outyielded their haploid parents as well as some 4x cultivars.

Results on the extraction of haploids from varieties adapted to Italian conditions are reported in this paper.

MATERIALS AND METHODS

Four varieties widely grown in Italy (Desiree, Jaerla, Primura and Sirtema) were crossed with pollinator S. phureja, clone PI 1.22. All crosses were done using the technique described by Peloquin & Hougas (1958). Seedlings from the crosses were grown in the glasshouse and root-tip chromosomes were counted.

RESULTS AND DISCUSSION

The results of the pollination, in terms of fruit set, seeds, seedlings, haploid frequency and haploids per 100 fruits are presented in Table 1.

A total of 863 pollinations were made which resulted in 286 fruits, 329 seeds and 250 seedlings. More than 14% of the seedlings were found to be haploids with an average of 12.1 haploids per 100 fruit.

The ability of the clone PI 1.22 to induce haploids with the four parents used was confirmed.

INTRODUCTION

In the Netherlands most potato breeders grow seedlings in the glasshouse in pots. The heaviest tuber per plant is harvested as seed for the first clonal generation. The seed tubers are small and moreover weight varies enormously within populations. As seed tuber weight affects the phenotype of the first year clones, the variation in weight has a negative effect on selection efficiency. This has been confirmed by the results of Blomquist and Lauer (1962), Swiezynski (1968), Brown et al. (1984) and Maris (pers. comm.). These workers all found that plants obtained from heavier seed tubers had better chances in plant selection but none could specify a weight interval at which tuber weight plays a less decisive role in the selection of first year clones.

Research on this selection problem will be presented in this paper.

MATERIALS AND METHODS

The influence of weight of seed tubers on the phenotype of first year clones is being studied in two similar 2-year experiments. Experiment 1 started in 1984 and ended in 1985 whilst experiment 2 started in 1985 and will be concluded in 1986. During the first experimental year, first year clones were grown in the field. Each genotype was represented by six plants (experiment 1) and eight plants (experiment 2), obtained from tubers differing in weight. For experiment 1 tuber weight ranged between 1.6–80.3 g and for experiment 2 between 1.5–77.5 g.

INTRODUCTION

From the title, the reader of this chapter may expect to get the formula for the Dutch breeders' success. However, there never has been a specific breeding strategy formulated by the fundamental breeders although, in the past, possibilities for future work have been published (Dorst 1963; Anon. 1980). The viewpoint put forward in this paper is that of a practical breeder, and so runs the risk of being one-sided.

The paper has been divided into two sections:

1. a. the past, because this has led to current achievements,

2. b. the future, what is desirable and what new opportunities will there be?

THE PAST

Potato breeding in the Netherlands has now been practised for about a century. In that period breeding has developed into a fully-fledged profession. Nevertheless, the basic principles of breeding are still the same - creating variation and selection.

The beginning in the Netherlands

There was no obvious date when breeding commenced. In several countries, during the whole of the last century, naturally produced true seed was grown to prevent degeneration. This slowly evolved into purposeful crossing (Sneep 1968). In the Netherlands the work of G. Veenhuizen in 1888 is considered to be the start of potato breeding. His great influence is described by De Haan (1958).

INTRODUCTION

Quantitative resistance to diseases of potato varieties is commonly expressed as a value on a 1–9 scale in publications by testing authorities, in advisory and technical literature and in information supplied by breeders prior to official trials. It is an effective means of summarizing a wide range of characteristics, whereby a high value indicates a desirable and a low value an undesirable quality. The values indicate relative susceptibility and may be interpreted by comparing values with those of known varieties.

Valid comparisons between 1–9 values obtained by different organisations can only be made with a knowledge of:–

1. (1) the experimental procedures used,

2. (2) the method of calculating 1–9 values from the experimental data and

3. (3) the control varieties used for these calculations.

EXPERIMENTAL PROCEDURES

A summary of the procedures used at the National Institute of Agricultural Botany (NIAB) is as follows:–

Late blight (Phytophthora infestans) in foliage: Glasshouse plants are sprayed with a suspension of zoospores and incubated at 15°C and 100% rh with artificial light (Saunders 1968). The proportion of the leaf surface affected is assessed after 7 days.

Late blight in tubers: Newly harvested tubers are sprayed with a suspension of zoospores and incubated at 15°C and 100% rh in the dark (Stewart et al. 1983). The proportion of the tuber surface affected is assessed after 10 days.

Common scab (Streptomyces spp.): A soil moisture deficit to induce common scab is created by erecting polythene tunnels to cover field plots immediately prior to tuber initiation (Jellis 1975).

Progress towards breeding potato varieties resistant to Globodera rostochiensis (Woll.) and G. pallida (Stone) has advanced considerably over the past decade. Recent developments at the Scottish Crop Research Institute (SCRI) in testing progeny with quantitatively inherited resistance, i.e. those derived from Solanum vernei (Bitt et Wittm.) and S. tuberosum ssp. andigena (Juz et Buk.) CPC 2802, have allowed genotypic parental values for resistance to G. pallida to be estimated (Phillips & Dale 1982). Data from such seedling progeny tests also allow the plant breeder to identify the most resistant progenies for further assessment and subsequent selection.

In parallel with improved progeny and parental selection for quantitative resistance to G. pallida has been the selective breeding and test crossing of parental material to produce parents which are triplex or quadruplex for the major gene H1, which confers resistance to G. rostochiensis (Ro1 and Ro4). The use of such parents guarantees that all members of derived progenies are resistant, eliminating the need for routine clonal testing for the H1 gene and thus releasing resources (Mackay, this volume).

After sowing the selected progenies, clones produced are tested for resistance to both G. rostochiensis and G. pallida in the third clonal year using closed containers (see also Lacey et al., this volume). Within these tests resistant, partially resistant and susceptible standard controls are used, allowing easy reliable comparisons both within and between tests and also between laboratory, glasshouse and field tests.

Although potato breeding strategy must take account of new developments such as in vitro tissue culture, somatic fusion, dihaploids and true potato seed, further improvements still need to be made in the classical techniques, including the selection of parents, the choice of crosses, and hybridization and selection of resulting progeny. This report discusses selection in early generations.

It is well known that the identification of desirable genotypes is complicated by variability in yield and other characters due to such factors as year, location, agronomy and experimental method. These environmental factors affect all stages in the breeding programme but are a particular problem in the early stages when there is no replication and only 1–10 plants per clone; nevertheless 99% of the initial population is discarded at this stage in the Omsk programme.

The influence of year, agronomy, variety and plant spacing on yield, starch content and protein content of individual tubers has been studied. Overall, genotype had a greater influence on starch content than on yield and protein content. There was a close correlation between protein content and fertilizer treatment. The correlations between first tuber generation, or A clones, and second tuber generation, or B clones, varied between r = -0.014 and r = +0.301 depending on the character (Table 1).

Wild potato species are an immense source of genetic variability which has been relatively little used. Many diploid and tetraploid wild species are compatible with diploid clones of Solanum phureja cultivated on the Andes of South America. The resulting hybrids can have various levels of ploidy:

1. a) diploid, 2n = 24 (wild diploid x cultivated diploid)

2. b) triploid, 2n = 36 (wild tetraploid x cultivated diploid)

3. c) tetraploid, 2n = 48 (wild tetraploid x cultivated diploid producing 2n gametes.

Gene transfer from wild to cultivated clones in the case of a) diploid and b) triploid hybrids is assured by utilizing them as male parents, which may produce 2n pollen grains by First division restitution or Second division restitution in the meiotic process, in order to fertilize tetraploid cultivars or clones of S. tuberosum ssp. tuberosum or spp. andigena. In the case c), in which the hybrids produced are tetraploid, gene transfer is more simple since the hybrids can be used either as males or females in backcrosses to tuberosum or andigena tetraploid clones.

By this method, hybrids were obtained from crosses between S. phureja clones and 12 species (S. acaule, S. andreanum, S. boliviense, S. bulbocastanum, S. chacoense, S. Colombianum, S. microdontum, S. sanctaerosae, S. sogarandinum, S. stoloniferum, S. toralapanum and S. vernei) belonging to seven taxonomic series (Figure 1).