For more than ninety years the two schools of genetics, the Mendelians and the biometricians, have been scientific rivals. The conflict polarized into the relative merits of single­gene resistance and poly­gene resistance. At present the argument still continues, but single­gene resistances firmly dominate plant breeding. However, the temporary nature of vertical resistances makes crops more susceptible to pests.

Before 1900, the science of genetics consisted of biometrics (Greek for 'life measurements'). Biometricians look at inherited characteristics as expressed on a quantitative basis, i.e. genetically inherited characteristics could be present in varying degrees. For example, a susceptible plant crossed with a resistant plant would produce progeny with every degree of susceptibility, ranging from the one extreme of minimum resistance, to the other extreme of maximum resistance.
However, the recognition of Gregor Mendel's laws of inheritance in 1900 drastically changed the science of genetics. Mendel's laws look at inherited characteristics as expressed on a qualitative basis. Genetically inherited characteristics are either present or absent, with no intermediates. For example, a white bean crossed with a black bean would produce only black and white beans, but no beans with degrees of greyness. Mendel's laws govern the ratio of black beans to white beans. The recognition of Mendel's laws marked the beginning of a conflict between the Mendelians and the biometricians on whether genetically inherited characteristics are based on one single gene or on polygenes.
The Mendelians had science on their side. The discovery of chromosomes made Mendel's discoveries very profound and convincing. The term 'gene' was coined to describe the unit of inheritance. Mendelian genetics had become a scientific bandwagon. Nevertheless, finding a single­gene characteristic that was of any economic importance in plants was difficult. All the characteristics of interest to plant breeders appeared to be quantitative in their inheritance, not qualitative.
However, in 1905 a British scientist, R.H. Biffen, discovered that wheat resistance to the rust disease was controlled by a single gene. Suddenly, the Mendelian school of geneticists discovered that genetically controlled characteristics that were of economic importance indeed exist. An explosion of research followed. As a result, resistance of many plant diseases was shown to be controlled by single genes. This was the resistance that is now called 'vertical resistance' (see box).
It was also shown that the quantitative characteristics valued by the biometricians did not conflict with Mendel's laws. Instead of being controlled by single genes, these quantitative characteristics were controlled by many genes. By the time the conflict with the biometricians was resolved, the Mendelians, focusing on single­gene inheritance were in complete control of plant breeding.
The single­gene resistances have both advantages and disadvantages. The advantages are complete protection against the parasite in question, and compatibility with breeding for wide climatic adaptation. These characteristics are attractive to large, centralized breeding institutes since these institutes target large areas and thus broad adaptability in their breeding material.
The main disadvantage of vertical resistance is its temporary nature, since it breaks down to new strains of the parasite. Other disadvantages include a loss of horizontal resistance while breeding for vertical resistance, and the fact that single genes for resistance cannot always be found. Thus, it has appeared impossible to breed for vertical resistance to some species of crop parasites, including many of the insect pests of crops. Moreover, vertical resistance has been misused in agriculture. Plant breeders have employed vertical resistance in uniform crop varieties, in which every host individual in one cultivar has the resistant gene.
On the other hand, scientific acceptance of horizontal resistance began slowly but many scientists still doubt the value of horizontal resistance, and some even its very existence. The reasons are that horizontal resistance is often difficult to measure, is dependent on environmental effects, and is often partial.

Horizontal and vertical resistance explained Infection is defined as the contact made by an individual parasite with an individual host for the purpose of parasitism. There are two kinds of infection: a parasite arrives from outside its host (allo­infection) or a parasite originates on (or in) its host (auto­infection). There are also two kinds of host resistance to crop parasites: vertical and horizontal. The inheritance of vertical resistance is controlled by single genes that are part of a gene­for­gene relationship, wherein for every host resistant gene, there is a corresponding infecting gene from a parasite. This relationship is an approximate botanical equivalent of the antibodies and antigens in mammals, except that the resistance is inherited and is functional before infection occurs. Vertical resistance can only control allo­infection. Auto­infection can be controlled by horizontal resistance only. In a wild plant pathosystem (i.e. the interaction between a host and pathogens) each host individual may have several vertical resistance genes, which constitute a biochemical lock, and each parasite individual may have several parasitism genes, which constitute a biochemical key. When a parasite allo­infects a host, its key either does or does not fit the lock of that host. When there is a mixture of many different locks and keys, most allo­infections are non­matching infections, and the growth of the epidemic (or infestation) is reduced very considerably. If every door in the town has the same lock, and every household has the same key that fits every lock, the system of locking will be ruined by uniformity. This is why vertical resistance is temporary resistance in agriculture. When a matching strain of the parasite appears, the resistance fails in every host individual of that crop and, shortly afterwards, of that entire cultivar. Horizontal resistance functions equally against most strains of the parasite. Consequently, it cannot fail to the extent that vertical resistance fails. It is durable resistance. Horizontal resistance is the resistance that invariably occurs in the absence of vertical resistance, or after a vertical resistance has been matched. Auto­infection by an asexually reproducing parasite is a matching infection, and the consequences of a matching infection, including all auto­infections, can be controlled by horizontal resistance only. Horizontal resistance tends to be lost when crops are bred for vertical resistance, or when they are bred under the protection of crop protection chemicals. Consequently, most modern cultivars have considerably less horizontal resistance than the cultivars of 1900s.

Plant breeding for horizontal resistance
During the 1960s, many plant breeders began to doubt the profitability of breeding for vertical resistance. The commercial life of most vertically resistant cultivars was too short to justify the amount of breeding work. This attitude, combined with the development of improved crop protection chemicals, and the investments of chemical industries in breeding, led to a gradual abandonment of resistance breeding in favour of crop protection by chemicals.
At present, the world spends about US$ 9 billion annually on crop protection chemicals. Despite this, pre­harvest crop losses due to pests and diseases are estimated at 24 per cent. In food crops alone, these losses are enough to feed about one billion people.
The only effective means of overcoming corporate and scientific opposition to horizontal resistance is to make plant breeding as public and as widespread as possible. Fortunately, breeding crops for horizontal resistance is so easy that it can be undertaken as a group activity. These breeding group could be farmers, hobby gardeners, green activists, environmentalists, or university students.
As a first step, a breeding group must start with a reasonably wide genetic base of susceptible plants. It is not necessary to find a good source of resistance, as when breeding for vertical resistance. Transgressive segregation within a population of susceptible plants will usually accumulate all the horizontal resistance needed. Should it not, merely widening the original genetic base will probably suffice.
A second step is the use of recurrent mass selection as a breeding method. This means that about 10­20 original parents, high quality modern cultivars but also land races, are cross­pollinated in all combinations. The progeny should total some thousands of individuals that are screened for resistance by being cultivated without any crop protection chemicals. The majority of this early screening population dies, and the parasites do most of the work of screening. The survivors become the parents of the next generation. This process is repeated until enough horizontal resistance is accumulated. Usually, a maximum of about 10 to 15 generations (which take 10 to 15 years in temperate climates, but less if more seasons per year could be realized) of recurrent mass selection will produce high levels of horizontal resistance to all locally important parasites.
Recurrent mass selection must be performed 'on­site', i.e. (a) in the area of future cultivation; (b) in the time of year of future cultivation; and (c) according to the farming system of future cultivation. This will produce new cultivars that are in balance with the local agro­ecosystem.
Breeders for horizontal resistance need to follow three simple rules when conducting recurrent mass selection. First, they should select individual plants on the basis of high yield, because the least susceptible plants yield the most. All measurements are relative: the best yielding plants are kept, even though their yields may be very low in the early screening generations.
Second, breeders should use simple techniques to introduce a parasite to each individual plant to ensure that all plants in the screening population are infected or infested with parasites. This will guarantee that the high yields are due to resistance, and not to chance escape from infection or infestation.
Third, a gene­for­gene relationship can occur, i.e. each gene of resistance in the host has a matching gene of parasitic ability in the parasite causing the plant to become susceptible to it. When this occurs, breeders should use a very simple technique called the 'one­pathotype' technique (a technique for ensuring that all vertical resistances are matched during the process of screening for horizontal resistance). This will ensure that the resistance is horizontal and not vertical.
Raoul A. Robinson

445 Provost Lane, Fergus, Ontario, Canada, N1M 2N3.
Phone (+519) 843 2355; E­mail raoulrob@sentex.net

For more detailed information on horizontal breeding see R.A. Robinson (1996), Return to Resistance: Breeding crops to reduce pesticide dependency. California, USA: AgAccess and Ottawa, Canada: IDRC Books.

Source
D. Pimental and H. Lehman (eds.) (1993), The Pesticide Question: Environment, economics, and ethics. New York: Routledge,
Chapman & Hall.