Mutation Breeding, Genetic Diversity and Crop Adaptation to Climate Change

Section 1.

Contribution of Crop Mutant Varieties to Food Security

1 World Food Supply: Problems and Prospects

J. Perry Gustafson¹ and Peter H. Raven

¹ Email: [email protected]

Abstract

The current world population of 7.6 billion is projected to reach 9.9 billion by 2050. The UN projects that agricultural output will need to increase by 70% simply to maintain current dietary standards, which does not include improving the diets of approximately 800 million malnourished people. Agricultural production increased at a rate insufficient to reach the goal set by the 2009 World Summit on Food Security to reduce by one half the number of malnourished people in the world by 2015. In spite of declining poverty rates, achieving this reduction in the number of malnourished people will be very difficult, as it is likely that the projected 2.3 billion additional people will be among the poorest of the poor. Food imports are expected to increase despite projected increased production, with many poor countries unable to afford those imports. Agriculture can improve sustainable world food production on the land currently under production and by doing so protect our fragile environment as much as possible.

Keywords: population growth • food supply • agriculture

1 Introduction

When our ancestors adopted crop agriculture, some 10,500 years ago, the entire global population amounted to about 1 million people, with only about 100,000 in all of Europe. It took until approximately 1810 for the world’s population to reach 1 billion people for the first time, and today it has grown to more than 7.6 billion, increasing by about 200,000 per day, and is estimated to reach 9.9 billion, an increase of 29%, by 2050 (PRB, 2018) and then perhaps 11 billion people by the end of the century, thus placing tremendous pressures on increasing world food production of all kinds. The United Nations (UN) projects that by 2050 agriculture will need to increase production by 70% (Alexandratos and Bruinsma, 2010) in order to maintain the same dietary standards that we have today, but the world would still have more than 800 million severely undernourished people and more than 100 million living near starvation. In 2018, world cereal production was estimated to be 2.56 billion tonnes, thus the world’s farmers will need to produce 3 billion tonnes by 2050 to meet the FAO projections – and according to Fischer et al. (2014) it is likely this can be achieved. As approximately 2.3 billion poor people could be added during that same period, it is just a rough estimate to give what increases will be needed. Since 1990, the number of undernourished people has risen to 815 million on a global scale but did actually decline from 900 million in 2000. Over the whole period from 1990 to 2010, world food production per capita increased by 12% (Barrett, 2010). Unfortunately, agricultural production was projected to increase by only about 1.3% per year during the period to 2030 (Mann, 1999). Fischer et al. (2014) placed it higher, at around 1.5%, for cereals. Therefore, it did not seem possible for agricultural output to come close to reaching the 2009 World Summit on Food Security goal of reducing by 50% the number of hungry and malnourished people in the world by 2015 (FAOSTAT, 2010). The World Bank (2008) estimated that in 1990 approximately 1.13 billion people worldwide were living in poverty, but 10 years later this figure was thought to have dropped to about 800 million. In spite of all our technology and agricultural improvements, in 2010 roughly 1 billion people still faced hunger on a daily basis. However, one must look very closely at the numbers involved. Some claim that there is no longer a major world food problem, as currently evidenced by the improvements in many developing countries, where the middle classes grow and there is a decrease in the percentage of poor people. This is true if we look at the percentages, but we should not look at percentages; instead, we need to look at the numbers, since there are about 1 billion impoverished people today when the world population is over 7.6 billion (14%), but when the world population reaches over 9 billion people, all projections show a decline in percentage of poor people.

Even with the projected increase in food production, food imports, on a world scale, will continue to increase; for example, wheat imports were projected to increase from 30 to 75 million tonnes by 2020 (Pingali and Rosegrant, 1998). However, the US Department of Agriculture (USDA) projected that increases in world wheat imports for 2018 would be about 170 million tonnes. Imports cannot solve world food problems within a ceiling of inadequate production, especially since only countries that are wealthy enough will be able to afford them.

In the past, as the human population grew, crops and domesticated animals provided a stable source of food; villages, towns and cities were formed and within them was developed the essence of what we consider civilization today. People could specialize as political or religious leaders, music makers, storytellers, historians and gradually into all of the myriad professions that we practise today. By about 5000 years ago, writing had been invented and history became recorded much more formally than earlier. What were once small groups of people became larger and more powerful, warring with one another for land and wealth. More than 200 million people have died in wars over the past two centuries, while our overall population grew from 1 billion to 7.6 billion. Humans are certainly the dominant species now and we often seem to fail to recognize our limits as we drain the world’s potentially sustainable resources and compete with one another for arable land and food supplies, wealth and ever-higher levels of consumption.

Plants, once domesticated, were continually improved by selecting seeds and individuals that produced the highest yields and at the same time were sustainable. In 2010, FAOSTAT data showed a phenomenal increase in world food production over the preceding 47 years to about 7.99 billion tonnes of total food production, including from both plant (6.90 billion tonnes) and animal (1.09 billion tonnes) sources, on basically the same amount of land (USDA-NASS, 2010). The dramatic increase in world crop production, which supplies feed for domestic animals and food for human consumption, has mainly come from improved crop cultivars, technology advances and improved management practices. Over the next 40 years, about 80% of the required increase in world food production was projected to result from yield increases (67%) and higher cropping intensities, i.e. better management (12%), with the remaining 21% coming from a minimal expansion of arable land under crop production (World Bank, 2008).

2 Requirements and Impediments

The main objectives in feeding the world’s population include, first, the task of increasing world food production, and improving dietary standards of the chronically undernourished, for the expanding population. Secondly, even if agriculture can accomplish the daunting task of increasing production, we will have to deal with the overwhelming task of improving and expanding the world’s infrastructure in order to distribute food equitably to all regions of the world to offset the current consequences of population growth and increasing world hunger. We will never see a lasting solution to the world food/hunger and poverty problem without a strong balance between food production and distribution – in other words, social justice. Thirdly, agriculture needs to accomplish these objectives with a minimum impact on the world’s biodiversity and our fragile environment. Fourthly, the increased food production needs to be accomplished without significantly expanding existing levels of land under cultivation. Approximately 11% of the world’s landmass is used for crop production, with an estimated 22% more used for pasture, most of which is natural grassland and manifestly unsustainable for crop production. No more than 10% of additional arable land is potentially available even for limited crop production (Bruinsma, 2003). The massive increase in food production between 1963 and 2005 was accomplished utilizing about the same amount of land as was under production in 2010 (USDA-NASS, 2010). For example, world grain yields more than doubled from 1.4 tonnes per hectare (t/ha) in 1961–1963 to 3.05 t/ha in 1997–1999, at the same time as the amount of land required for producing the increase in grain yield actually declined by approximately 56% (World Bank, 2008). About one-eighth of our necessary protein comes from the sea, but this supply does not appear elastic in relation to demand, and again, most of it is taken by wealthy countries.

There is clear scope for increases in world food production on existing agricultural land and these increases should be feasible utilizing existing and newly developed technology. It was estimated that in parts of South-east Asia, the average rice yields were only 60% of their average maximum climate-adjusted yields (Godfray et al., 2010). In addition, 11 countries were producing 37% of the total world wheat tonnage on predominantly rainfed production conditions, well under their attainable yield potential (World Bank, 2008). If farmers around the world were able to produce closer to their potential yield, the world’s production levels should significantly increase without any additional land being brought into cultivation. This increase in production should equal over 23% of the current world production. As has been shown in the past 50 years, any local or national improvement in cultivars and management will spill over into the rest of the world. The world needs to understand that agriculture is not a local or national industry and any changes that are made on a local level will have global impacts. On the other hand, the droughts in Australia and Argentina seem tied to global climate change, as does the irregularity of the monsoon season in India. Therefore, food production increases will surely be subject to a number of additional constraints yet to be defined.

Even though this discussion indicates the potential for major increases in world agricultural production on existing cultivated land, based on existing and newly developed technology, there are at least three impediments that could limit increased world food production.

1. The technology and management improvements required for advancing yield might not be accessible or applicable to all crops, regions and farmers of the world.

2. Advanced technology and management inputs in agriculture could spread into areas of the world where they could accelerate environmental problems and have an adverse impact on biodiversity.

3. The understanding of agriculture by the public and by farmers certainly needs to be vastly improved for productivity to be increased, because we need a cooperative environment where agriculture can grow as increased numbers of people move into cities and tend to place obstacles in the way of improving crop production or to pay insufficiently to support those who raise our food.

Intensifying agricultural technology on existing lands, therefore, has had and will continue to have a major role in preserving biodiversity and maintaining the sustainability of our fragile global environment overall.

3 Technology for the Creation and Deployment of Diversity

One vital need to eliminate hunger and poverty involves the preservation of sufficient genetic diversity in cultivated plants and their relatives to ensure that breeders have the capacity to create cultivars capable of resisting biotic and abiotic stresses and adapting to new environmental conditions. Existing, improved and newly developed biotechnological tools alongside traditional technology will play major roles in improving world food production, in the same manner as did the Green Revolution that occurred from the 1960s through the 1980s. This will amount to what Conway (1999) termed ‘the doubly green revolution’. The main problem will be the degree to which individual countries, industrialized and developing, can manage existing and new technologies to adapt and improve their production with advantageous changes in agronomic practices and production costs without any adverse effects on the world’s environment and biodiversity.

The domestication and development of crops over the past 10,500 years has involved progeny selection from the most productive individuals based on their phenotypes. Existing and new biotechnologies are taking plant improvement to new heights with the potential of greatly improving food quality and production. Existing and new agricultural technologies are composed of individual systems that involve both public and private breeders’ plant selection programmes and involve many technologies, including the following.

1. Tissue culture, in which plants can be broken down into cell suspensions that are manipulated followed by regenerating plants, bypasses the traditional approaches to seed production.

2. The utilization of anther culture coupled with various chromosome-doubling techniques can successfully create double haploid populations, greatly reducing the time required to produce commercial cultivars.

3. Modern approaches to mutation technology have been and will continue to be successful in creating additional genetic variation necessary for crop improvement programmes.

4. More recently, the utilization of molecular marker-assisted selection, where various types of DNA marker systems are linked to traditionally difficult-to-screen value-added traits of interest, have already been successfully used in cultivar improvement programmes (for an excellent review see Tester and Langridge, 2010). Among the techniques most commonly being used today are: restriction fragment length polymorphism (RFLP) and amplified fragment length polymorphism (AFLP), which at the present time are not being widely used for marker-assisted selection; simple sequence repeat or microsatellite repeat (SSR); single nucleotide polymorphism (SNP); and diversity array technologies (DArT).

5. The application of genome-wide selection is mainly used for selection involving quantitative traits in both animals and plants (Goddard and Hayes, 2009; Mayor and Bernardo, 2009). However, genome-wide selection requires the availability of a massive number of very cheaply available markers (up to 20 markers per centimorgan (cM, the unit for measuring genetic linkage)) or a high-density chip specifically for the species being manipulated, which canan be very expensive technology. Also, when selecting on a genome-wide scale one has to consider the presence of considerable linkage drag.

6. The application of genomic sequencing of individual plant genomes to expose the location and potential function of the entire genetic composition of an organism can be used in conjunction with other technologies to assist cultivar improvement programmes.

7. Plant transformation technology creates genetically modified organisms (GMOs). This involves bypassing the sexual process to transfer genes from one organism to another and has already been successful in several crops.

A brilliant new technique for improving the agronomic characteristics of plants has emerged in the past two decades: the application of clustered regularly interspaced short palindromic repeats (CRISPR) and CRISPR-associated (Cas) systems (reviewed by Bomgardner, 2017), which is currently showing considerable potential applications in biotechnology crop improvements. The identification of a target site within a target gene of a plant species is quite straightforward using CRISPR technology, which requires the editing components to be delivered to plant cells. Using the current technology, it is common to introduce the selected genes as transgenes, but CRISPR methods are more direct. Their use is spreading rapidly and will clearly bring great benefit to the characteristics in the future, a tremendously important advance given the major changes in climate that we are facing over the decades to come. The ability to produce targeted mutations in any plant gene is extremely valuable in manipulating specific genes within a crop which are likely to improve its most critical features.

Most of the traditional and newly developed technologies have been adapted to work efficiently in a more land- and labour-intensive form of agriculture improvement. It is clear that traditional organic farming applications are not capable of producing enough to feed or improving dietary standards of the existing world population, let alone the projected increase to 9.9 billion people by 2050. It has been estimated that a world population of only 4 billion people could be sustained if organic nitrogen farming systems were in place on a global scale (Buringh and van Heemst, 1977; Smil, 2001, 2004; Conner, 2008), though requiring significantly more land under production to generate the required organic nutrients. All things considered, we are not really sure how many people the world can support sustainably, but we clearly need to develop sustainable, productive agriculture to the extent that we can while working with other factors leading to sustainability at the same time.

4 Discussion

Existing data indicates that agriculture is capable of feeding the projected increased world population on approximately the same amount of land that is currently under production (World Bank, 2008). However, it will take all of the available technologies and skills of plant and animal breeders and agronomists, coupled with the actions of farmers, to achieve the desired goal of eliminating world hunger. In addition, integrated pest management (IPM), water management, precision farming, limiting chemical input and many other techniques must be wisely applied and improved simultaneously to maximize yield from existing farmland. Only the coordinated application of all these techniques will improve the productivity of the lands currently cultivated to the degree needed.

Significant progress has been made over the past several years in advancing our knowledge of biology, which is being applied to technology adaptable to improving agricultural production. We do not know the direction current research will take, but we can assume that any commercial application is going to be determined by world economic and social factors. The private sector and many funding agencies are increasingly addressing the agricultural needs of developing countries. That a ‘back to nature’ or the ‘pure organic’ approach can feed the world’s people is a theory that does not take into account the current world population and the scale of human suffering from malnutrition and starvation. Embracing social justice is the only way that people can really survive and prosper and we need to utilize all of our resources to accomplish the goal of feeding humanity.

All crops will continue to be improved by traditional and biotechnological approaches to advance their yield potential. Building adaptable gene complexes into crop varieties for the future is something that we must do, especially in the face of global climate change and the world’s increasing population. This process will require a much larger number of cultivars, with different genetic backgrounds, to be spread around the world than in the past. In future crop development, several factors seem especially important.

1. We need to understand the characterization of the genome structure, gene function and regulation, and evolution at macro- and micro-geographical scales of all crops and animals.

2. We need to combine single- and multi-locus value-added traits to produce a higher degree of cultivar development.

3. Crop genetic systems must be analysed to determine the genetic flexibility of various species in diverse ecological contexts, according to their breeding systems, mutation rates, genome recombination properties, genomic distribution and function of structural genes (primarily abiotic and biotic stress genes).

4. We need to characterize the interface between developing agricultural ecological dynamics and adaptive ecosystems in order to manipulate genome composition and limit the potential for gene contamination.

In the past, when modern agriculture competed with the traditional subsistence forms of agriculture, local landrace cultivars were often discarded in favour of the new high-yielding cultivars. Recently, massive efforts have been undertaken to preserve crop and animal diversity, which has resulted in the characterization of more old and new landrace diversity of all kinds, and these are being more widely used in agriculture and breeding programmes than 50 years ago. Extraordinary efforts are being made in this area, but much crop diversity is being lost at the same time. National and international seed banks are and will continue to be critically important to agriculture and the maintenance of the world’s biodiversity. The continued long-term health of world food production is one of the foundations to world security. The stable future of humanity, our environment and our biodiversity are intimately tied to the improvement of crop production. Adequately feeding the world’s population is clearly one of the most important challenges facing the world today and in the future.

At the same time, and as rapidly as we can, we need to achieve a global population level that is sustainable, an achievement that will clearly depend in a major way on empowering women throughout the world. Efforts to do this are increasing in some areas, but they need to be greatly strengthened so that women everywhere can gain the right to control their own destinies, make whatever contributions they can to our common good and determine their own reproductive decisions. We need to achieve just and sustainable levels of consumption worldwide; five planets would be necessary if everyone lived at the current level of consumption in the USA, while at the same time many of the poorest nations are entering what Mathis Wackernagel, founder of the Global Footprint Network, has termed an ‘environmental poverty trap’ as a condition that will deny them the resources to import food. There is little evidence that wealthy nations will step in to feed the countries of sub-Saharan Africa, for example, which makes the concept of producing enough food for the whole world somewhat of an illusion. We need to care enough to stop global climate change if agriculture is to stay close to its present levels of productivity, let alone improve them. Overall, a world in which national greed is dominating as a policy can never be a world where everyone is adequately fed, regardless of how much food might be produced. We need social justice and love for one another if we are ever going to achieve such a goal.

References

Alexandratos, N. and Bruinsma, J. (2010) World Agriculture Towards 2030/50. Agricultural Development Economics Division, FAO, Rome. Revised and updated 2012 (ESA Working Paper no. 12-03).

Barrett, C.B. (2010) Measuring food insecurity. Science 327, 825–828.

Bomgardner, M.M. (2017) CRISPR: A new toolbox for better crops. Chemical & Engineering News 95 (24), 30–34.

Bruinsma, J. (ed.) (2003) World Agriculture: Towards 2015–2030 An FAO Perspective. Earthscan Publications, London.

Buringh, P. and van Heemst, H.D.J. (1977) An estimation of world food production based on labour-oriented agriculture. Centre for World Food Market Research, Amsterdam.

Conner, D.J. (2008) Organic agriculture cannot feed the world. Field Crops Research 106, 187–190.

Conway, G. (1999) The doubly green revolution: food for all in the twenty-first century. Cornell University Press, Ithaca, New York.

FAOSTAT (2010). Food and Agriculture Data. Available at http://www.fao.org/faostat/en/ (accessed 2 September 2021).

Fischer, R.A., Byerlee, D. and Edmeades, G.O. (2014) Crop yield and global food security: will yield increase continue to feed the world? ACIAR Monograph 158. Australian Centre for International Research, Canberra. Available at: http://aciar.gov.au/publication/mn158 (accessed 2 September 2021).

Goddard, M.E. and Hayes, S.J. (2009) Mapping genes for complex traits in domestic animals and their use in breeding programmes. Nature Reviews Genetics 10, 381–391.

Godfray, H.C. et al. (2010) Food security: The challenge of feeding 9 billion people. Science 327, 812–818.

Mann, C.C. (1999) Future food: crop scientists seek a new revolution. Science 283, 310–314.

Mayor, P.J. and Bernardo, R. (2009) Genomewide selection and marker-assisted recurrent selection in doubled haploid versus F2 populations. Crop Science 49, 1719–1725.

Pingali, P.L. and Rosegrant, M.W. (1998) Supplying wheat for Asia’s increasingly westernized diets. American Journal of Agricultural Economics 80, 954–959.

PRB (2018) 2018 World Population Data Sheet. Population Reference Bureau, Washington, DC. Available at: www.worldpopdata.org (accessed 24 March 2021)

Smil, V. (2001) Feeding the World: a Challenge for the Twenty-first Century. MIT Press, Cambridge, Massachusetts.

Smil, V. (2004) Enriching the Earth: Fritz Haber, Carl Bosch, and the Transformation of World Food Production. MIT Press, Cambridge, Massachusetts.

Tester, M. and Langridge, P. (2010) Breeding technologies to increase crop production in a changing world. Science 327, 818–822.

USDA-NASS (2010) United States Department of Agriculture – National Agricultural Statistics Service. Available at https://www.nass.usda.gov (accessed 2 September 2021).

World Bank (2008) World Development Report: Agriculture for Development. The World Bank, Washington, DC.

2 Scandinavian Mutation Research During the Past 90 Years – a Historical Review

Udda Lundqvist¹

¹Email: [email protected]

This chapter is an updated version of a previously published article on the history of Scandinavian mutation research (Hereditas 151, 123–131). It is published here with the permission of the author and of Hereditas.

Abstract

In 1928, the Swedish geneticists Herman Nilsson-Ehle and Åke Gustafsson started to act on their own ideas with the first experiments with induced mutations using diploid barley. They started with X-rays and UV irradiation. Very soon the first chlorophyll mutations were obtained and followed by the first ‘vital’ mutations Erectoides (ert) (Franckowiak and Lundqvist, 2001). Several other valuable mutations were identified as early maturity, high yielding, lodging resistant and characters with altered plant architecture. The experiments expanded to include other different types of irradiation, followed by chemical mutagenesis starting with mustard gas and concluding with sodium azide. The research brought a wealth of observations of general biological importance, such as the physiological effects of radiation as well as the difference in the mutation spectrum with respect to mutagens.

This research was non-commercial, even if some mutants have become of important agronomic value. It peaked in activity during the 1950s to 1980s and, throughout, barley was the main experimental crop. About 12,000 different morphological and physiological mutants with a very broad phenotypic diversity were brought together and are incorporated in the Nordic Genetic Resource Centre (NordGen), Sweden. Several important mutant groups have been analysed in more detail genetically, with regard to mutagen specificity and gene cloning. These are: (i) early maturity mutants (Praematurum); (ii) six-rowed and intermedium-spike mutants; (iii) mutants affecting surface wax coating (Eceriferum); and (iv) mutants affecting rachis spike density (Erectoides). Some of these groups are presented in more detail in this review.

Once work with induction of mutations began, it was evident that mutations should regularly be included in breeding programmes of crop plants. In Sweden, direct X-ray induced macro-mutants have been successfully released as cultivars, some of them having been used in combination breeding. Their importance for breeding is discussed in more detail.

Keywords: mutation • barley • genetics • breeding

1 Introduction

Swedish research on induced mutations in barley started in 1928 on a small scale at Svalöf, Sweden, initiated by the eminent Swedish geneticists Herman Nilsson-Ehle and Åke Gustafsson (Fig. 2.1). Although L.J. Stadler in the USA also published data on induced mutations in barley, he described them with much pessimism, stating that they would be of no use for plant breeding (Stadler, 1928). Nilsson-Ehle and Gustafsson did not share this pessimism and at Gustafsson’s suggestion experiments were initiated in barley (Hordeum vulgare) with induced mutations. The first treatments with X-rays and ultraviolet irradiation commenced using the Svalöf cultivar ‘Gull’, which was the most common barley cultivar grown in Sweden at that time. Different pre-treatments were also tested, since it was known that mutation frequency increased if seeds were soaked in water before irradiation (Stadler, 1928). The first chromosome aberrations were observed and the first phenotypic changes in the seedlings, chlorophyll mutations, occurred. At that time distinct categories albina, viridis and xantha (Fig. 2.2) were established, and the rare two-coloured striped and zoned phenotypes were recognized. These chlorophyll mutations were always the first indication of treatment success (Gustafsson, 1938, 1940) and their abundance served as the standard method for measuring the induced mutagenic effects.

Fig. 2.1. The eminent Swedish geneticists for mutation research and breeding.

Fig. 2.2. The three distinct categories of chlorophyll mutations.

In the mid-1930s the first viable mutations appeared and it was possible to distinguish two subgroups: ‘Morphological’ and ‘Physiological’ mutations. The most common group at that time was the Erectoides mutants characterized by a compact or dense spike. Morphologically they resemble the erectum barleys, in comparison with the normal nutans spikes in most of the barley cultivars. In the following years, several of the mutants that were produced were considered very valuable, with characteristics such as high yield, straw stiffness, straw length, early maturity, tillering capacity, changes in spikes, kernels and awns, changed pigmentation and so on (Gustafsson, 1941, 1947). These results from all the early experiments looked so promising, even for plant breeding, that in 1940 the Swedish Seed Association at Svalöf started to sponsor this new research with funding from the Swedish milling industry. This made it possible to extend the experiments considerably. It was possible to integrate theoretical and practical results. In 1948, the Wallenberg Foundation incorporated these mutation activities in their research programme and permitted Gustafsson to gather around himself a group of specialists to carry on the research work with a wider perspective. In 1953, at the instigation of the Swedish Government, the ‘Group for Theoretical and Applied Mutation Research’ (Fig. 2.3) was established with the aim of studying basic research problems in order to influence and improve the methods for breeding of cultivated plants. The Agricultural Research Council provided funding for most of the Mutation Group’s scientific activities approved by the Swedish Parliament (Gustafsson, 1954). Its most intensive activities were during the 1950s, 1960s, 1970s and 1980s.

Fig. 2.3. The Swedish group for Theoretical and Applied Mutation Research (mid-1950s).

2 Materials and Methods

X-irradiation of dry seeds was the standard method for studying the mutation process, but soon other types of irradiation such as γ-rays for sparsely ionizing radiation, neutrons, electrons, protons for densely ionizing radiation, then α-rays from radon, and β-rays from phosphorus 32 and sulfur 35 were included in the experiments. Applications of treatments with different soaking times of the seeds, both before and after irradiation, were also studied. Not only was the water content of the seeds an important trait in relation to radiation sensitivity but so also were different environmental conditions. When comparing the two irradiation types, sparsely versus densely irradiated, it can be summarized as follows: seeds are 20–30 times more sensitive to neutrons than to X-rays; germinating seeds are two to three times more sensitive to neutrons than to X-rays; neutrons are approximately 10 times as effective as X-rays in producing chromosome disturbances and 50–100 times more effective in increasing the mutation rate in the second generation; and neutrons produce relatively more chlorophyll mutations than X-rays (Ehrenberg et al., 1952, 1959; Lundqvist, 1992).

Already in the mid-1940s, chemical mutagenesis had entered the scene and became included in experiments together with irradiation. The idea was to influence not only the mutation rate but also morphological and physiological types of mutations. The first to be applied were colchicine and mustard gas, followed by epoxides and epimines, most importantly ethylene oxide and ethylene imine, in addition to many different compounds such as purines and purine derivatives and various alkylating and oxidizing compounds. Then more complex compounds were applied, such as alkane sulfonic esters with monofunctional and di-and trifunctional sulfonates, N-alkyl-N-nitrosamides, organic sulfates and sulfonates, most commonly ethyl methanesulfonate (EMS), nitroso compounds, purine and acridine derivates and many other chemical mutagens. Finally, in the mid-1970s the inorganic chemical mutagen sodium azide was introduced and it became the most popular mutagen for the isolation of viable mutants for practical purposes in Swedish experiments. The mutation frequency in these experiments increased rapidly up to 80%; they were 20 times more effective than irradiation. Significant differences between the actions of ionizing radiation and chemical mutagens were demonstrated. It was shown that neutrons and sodium azide form two extremes: neutrons induce a relatively large number of chromosome changes, whereas sodium azide primarily causes gene mutations at the nucleotide level (Nilan et al., 1976). But differences in the mutation spectrum were also noticed, first with regard to chlorophyll mutations and later in some morphological and physiological mutation groups. Some gene loci reacted in different ways. The aim was actually to control the direction of mutagenesis. During all the years of working with mutation research, the following properties were always studied and determined in the different M generations.

1. The numbers of chromosome disturbances in germinating seeds were studied.

2. The numbers of germinating plants were counted in the M1 field generation.

3. The numbers of mature harvested plants in the M1 generation were registered.

4. The determination rate of chlorophyll mutations was studied in the seedling stage of the M2 generation. This part was very important, as it showed if the mutagenic treatment had been successful or not.

5. In the following generations all morphological and physiological mutants were studied and isolated.

This mutation research was non-commercial, even if some important mutants were used to develop commercial cultivars (Ehrenberg et al., 1956, 1961; Lundqvist, 1992, 2014).

3 The Swedish Collection of Barley Mutants

Genetic diversity in barley has been of great importance and has long been studied in great detail. This is important not only for plant breeding but also for investigations and localization studies of the barley genome. The Russian geneticist N.I. Vavilov felt it necessary to explore the total genetic diversity of crop plants throughout the world as well as diversity of related wild species (Vavilov, 1992). There will always be a large demand for a broad diversity, including genetically characterized mutants. These mutants will serve as basic material for all kinds of barley research and methodical work will sooner or later lead to positive results (Lundqvist,1986).

Over the years, a large collection of morphological and physiological mutations (about 12,000 different mutant alleles) with a very broad range of phenotypes were brought together and were genetically and agronomically studied. The collection consists of five main categories (Table 2.1). Groups A and B are divided into 12 sub-collections (Tables 2.1 and 2.2) (Lundqvist, 2005, 2014) and they are incorporated into the Nordic Genetic Resource Centre (NordGen), Sweden. Germination tests are conducted regularly, regeneration is done if necessary and the accessions are checked for their homozygosity. Most passport data are included as far as possible in the NordGen’s information system, SESTO. About 750 barley genetic stock descriptions (BGS) with all genes localized also with the Swedish mutants are published in the Barley Genetics Newsletter (BGN) and in the International Database for Barley Genes and Barley Genetic Stocks (Lundqvist et al., 2016; NordGen BGS, 2021). The Swedish mutants have been induced in different cultivars. In the USA, Jerome D. Franckowiak undertook a tremendous effort and by skilful crossing work he transferred many of the genes into a common background cv. ‘Bowman’, a two-row high-malting barley cultivar. He established about 1000 ‘Barley Near Isogenic Lines in Bowman’ (NIL); 60% of them include Swedish mutant genes. This NIL collection is very useful for linkage studies, assessment of specific marker genes and determination of linkage drag and has assisted gene transfer for all barley researchers worldwide. Due to this collection, several of the mutant genes have been cloned in recent decades (Druka et al., 2011).

Table 2.1. The five main categories of mutants.

Table 2.2. The 12 morphological and physiological barley sub-collections.

The Swedish collection is unique, since all alleles of investigated genes are conserved at the Nordic Genetic Resource Centre (NordGen), Sweden, and are available for all barley research and plant breeding. It is a major source for today’s gene mapping and valuable for molecular genetic analyses of cloned mutant genes. It also forms outstanding material for investigations within radiobiology, genecology, gene physiology, ultra-structural research, plant biochemistry, gene localizations, genetic fine mapping and molecular marker research. It serves as an important gene pool. During the years of the peaks of mutation research, about half of all mutants have been analysed genetically in greater or less detail, but they form only a minor part of the range of mutant characters. The mutant groups shown in Table 2.3 were studied and isolated in more detail genetically and also with regard to mutagen specificity. These studies have increased the knowledge of the mutation process and the architecture of different characters (Lundqvist, 1986, 1992, 2005, 2008, 2014). Some of them are presented and described in more detail in this review.

Table 2.3. Survey of the genetically investigated Scandinavian mutant groups.

3.1 Praematurum (early maturity) mutants

The demand for early-maturing cultivars has increased rapidly, because earliness has become an important goal for Swedish plant breeding and is an important feature under natural conditions. Farmers want an early crop to establish an effective crop rotation. Already in the 1940s, it was found that maturity in barley could easily be changed by using X-rays in either direction with both increased earliness and increased lateness. The time of heading was chosen as a safe character for screening induced early mutants, but early heading and early ripening are characters where environmental influences, especially climate conditions, may hamper a safe classification (Gustafsson, 1942; Gustafsson et al., 1960).

Over the years about 1250 different Praematurum mutants have been isolated and studied, using various mutagenic treatments. Several cultivars (‘Bonus’, ‘Foma’, ‘Kristina’, ‘Lotta’, ‘Semira’, ‘Frida’, ‘Golf’ and ‘Lina’) were used. Very soon the mutants could be grouped into three categories according to their heading and maturity time. We established three different classes of earliness, all compared with the mother cultivars, as follows: (i) drastically altered earliness of at least 1 week; (ii) medium increase of earliness of 3–5 days; and (iii) slightly modified earliness of 1–2 days (Gustafsson and Lundqvist, 1976; Lundqvist, 1992). Long-term studies made it possible to localize 195 mutants and distribute them to nine mat loci (Table 2.4). The different loci show in general quite distinct phenotypic characters. The mutations selected for earliness also change other properties of agricultural value. Significantly shorter straw with lower internode number is found in the extreme early mutant loci, mat-a, mat-b and mat-c. Mutants of locus mat-a are generally more resistant to lodging than mutants in locus mat-b and have generally a more reduced culm length. Thus, mutants of the Praematurum type may offer favourable materials for the selection of high-yielding and semi-dwarf types. Among these loci, the most dramatic is the Praematurum-a.8 (mat-a.8) mutant, derived as a one-step X-ray mutant, a drastically early mutant which heads 8–10 days earlier than its progenitor cultivar ‘Bonus’. It was approved and released as a commercial Swedish cultivar under the name ‘Mari’ in 1960 and was intended to replace early Swedish six-row cultivars. It was grown widely in Iceland and in the Mediterranean region and was also included in the International Maize and Wheat Improvement Centre (CYMMIT) barley breeding programme in Mexico (Hagberg, 1961; Sigurbjörnsson, 1976).

Table 2.4. Distribution of the early maturity mutants to the nine mat loci.

Not until the mid-1960s was it found that mat-a.8 had a special property that definitely distinguished it from the ‘Bonus’ parent cultivar, namely, a profound change in the photo- and thermo-period reaction, making it both heading and seed fertile at 8 h of daylight (short-day tolerant). During the 1960s, large phytotron experiments were carried out at the Stockholm Forest University under different photoperiod conditions to compare different mutants and cultivars (Dormling et al., 1966; Dormling and Gustafsson, 1969; Gustafsson et al., 1982). Later, when labour costs became too expensive, a darkening arrangement, using a special plastic tissue, was used in an ordinary glasshouse with natural light lasting for 8 h. This type of arrangement was applied over many years for identifying daylength-neutral mutants (Fig. 2.4). It was possible to distinguish three genotypic categories under the extreme short-day conditions of 8 h of light: (i) genotypes with complete early heading and good seed set; (ii) genotypes with incomplete and late heading and seed set; and (iii) genotypes that never headed but remained in a purely vegetative and often luxuriant state (Fig. 2.4). Regarding the mutants in mat-c and mat-e, they were characterized by less pronounced daylength neutrality compared with mat-a mutants. The mutants in all other mat loci were long-day adapted like the parent cultivars, being rather productive at 16–24 h of light, more or less infertile and late heading at 12 h, and not heading at all at 8 h of light (Gustafsson and Lundqvist, 1976). The induced early mutants tested derived from the whole spectrum of mutagens. There were rather more than expected short-day adapted mutants under sulfonate treatments, whereas the long-day adapted cases seemed to be in excess when ethylene imine was applied. Other observations indicated that sodium azide was less efficient in producing daylength-neutral mutants (Lundqvist, 1991, 1992). At the same time, Japanese researchers isolated short-day neutral mutants (ea8, eam8) allelic to the Praematurum-a (mat-a) gene (Yasuda, 1977).

Fig. 2.4. Heading mutants. Left: arrangement using a special plastic sheet to keep plants in the dark. Right: plants at maturity time showing mature mutant plants that never head (i.e. remain vegetative).

Due to the development of many molecular programmes, it was possible to clone two of the Praematurum genes: mat-a and mat-c. The Praematurum-a (mat-a) mutant has been identified as a homologue of the Arabidopsis thaliana circadian clock regulator EARLY FLOWERING 3 (ELF3). From 85 induced mat-a and 2 eam8 mutants, more than 20 different mat-a alleles were identified and these mutations predicted a defective ELF3 protein. Expression analysis of HvElf3 and HvGigantea in mutant and wild-type plants further characterized the flowering pathway leading to the early phenotype of the mat-a mutants (Zakhrabekova et al., 2012). The other early-maturity gene Praematurum-c (mat-c) was identified as a variant of the barley homologue of Antirrhinum majus CENTRORADIALIS (HvCEN) (Comadran et al., 2012). Of the 31 examined mat-c mutants, 29 different alleles were found (Comadran et al., 2012; Matyszczak, 2014) and led to the conclusion that variation in HvCEN was important in enabling geographical range extension. The mat-c mutants were also demonstrated to be alleles at the early maturity 6 (Eam6) locus (Comadran et al., 2012; Matyszczak, 2014).

3.2 The six-row (Hexastichon ) and Intermedium-spike mutant group

Barley is one of the oldest cultivated crops, and the number of rows of spikelets is a key character in inferring the origin of barley. After at least 100 years of discussion, it has been concluded that domesticated two-row barley is older than six-row barley, as is supported by archaeological specimens (Bothmer et al., 1995). The six-row (Hexastichon) and Intermedium-spike mutants affect the development of the lateral spikelets, with genetic interaction leading to synergistic enhancements. This research has given an insight into the rather complex genetics of kernel rows in barley. Normal two-row barley carries two spikelets that are on opposite sides of the spike and central spikelets with two reduced sterile lateral spikelets. Two-row barley is able to produce six-row barley in a single mutational step. These mutants have well developed central and lateral spikelets, are fully fertile and have long awns. All the 65 isolated cases have been localized to only one locus v1, renamed hex-v or vrs1, located in the long arm of chromosome 2H (Gustafsson et al., 1969; Lundqvist, 1992; NordGen BGS, 2021). But two-row barley may also produce mutants with spike development intermediate between the two-row and six-row states. These mutants have enlarged lateral spikelets that vary in characteristic ways with regard to awn and kernel development and fertility, not only among mutants, but also depending on environmental conditions. A total of 126 such Intermedium-spike mutants have been isolated and 83 of them have been localized to ten different int gene loci and studied in more detail (Table 2.5). All these mutations are recessive and show independent inheritance (Gustafsson and Lundqvist, 1980; Lundqvist and Lundqvist, 1988a). All Intermedium-spike mutants have been induced in the two-row barley cultivars ‘Bonus’, ‘Foma’, ‘Kristina’ and ‘Nordal’. Both radiation types and most of the used chemical mutagens were applied but no specific type of mutation could be associated with a certain type of mutagen (Lundqvist, 1992).

Table 2.5. Distribution of the Intermedium-spike mutants to the 10 gene loci.

During the 1970s Japanese researchers studied several such so-called ‘six-rowed’ mutants. All of them were recessive and they were assigned the gene symbols v2–v5 (Fig. 2.5), and were localized on the barley genetic map. In further studies at Svalöf, allelism tests were carried out between the Swedish int loci and the Japanese v genes. The results are summarized in Table 2.5 (Lundqvist, 1992). Due to newly developed molecular methods, all of these described vrs loci have been cloned. Six-rowed spike 1 encodes a homeodomain-leucine zipper I-class homeobox gene and its expression is strictly localized in the lateral spikelet primordial of immature spikes, suggesting that the Vrs1 protein suppresses development of lateral spikelets (Komatsuda et al., 2007). Vrs2 encodes a SHORT INTERNODES (SHI) transcriptional regulator which is expressed during barley inflorescence and shoot development (Youssef et al., 2017). Vrs3 encodes a putative histone Lys demethylase with a conserved zinc finger and Jumonji C and N domain and controls lateral spikelet development (Bull et al., 2017; Esse et al., 2017). Vrs4 also affects lateral spikelets and their fertility. This gene was cloned as an orthologue of the maize inflorescence architecture gene RAMOSA2 (HvRA2) which is associated with loss of spikelet determination (Koppolu et al., 2013). Vrs5 was cloned as an orthologue of the maize domestication gene TEOSINTE BRANCHED 1 (HvTB1); this gene modifies lateral spikelet development (Ramsay et al., 2011).

Fig. 2.5. Six-row and Intermedium-spike mutants.

3.3 Surface wax coating mutants: Eceriferum , Glossy (waxless)

Presence of a wax coating reduces evaporation of water from the plant and helps protect it against pathogens. Most surface wax mutants at the Eceriferum and Glossy loci affect the presence and type of epicuticular waxes on leaf blades, sheaths, culms and spikes. When the wax coating is completely absent, various organs appear as a bright, glossy green colour. Cooperation between Swedish and Danish researchers made this mutant type probably the best-known character complex of any cultivated plant. The mutants have been isolated, the genes localized, their influence on yield studied, loci have been mapped genetically and electronic microscopy and biochemical analyses have been done. Their reactions to various climates were studied in the phytotron. Phenotypically, three different organs of the barley plant were studied in regard to wax coating and composition. Five phenotypic categories were established: spike and leaf sheath; spike and leaf sheath partially; spike; leaf blade; spike, leaf sheath and leaf blade (Lundqvist and Wettstein, 1962; Lundqvist et al., 1968). A total of 1580 such Eceriferum mutants have been localized to 79 loci; 78 of them are recessive and one is dominant. Nearly all Eceriferum loci have been positioned on the barley genetic map. Seven types of mutagenic treatments have been applied and it is obvious that different loci show markedly different mutagen-specific reactions. Detailed analyses of mutation distribution led to the following results: (i) strong mutagenic differences between chemicals and ionizing radiation, especially neutrons; (ii) no significant differences among various kinds of organic chemicals; (iii) significant differences between organic chemicals and sodium azide; (iv) sodium azide differing strongly from X-rays and still more from neutrons; and (v) distinct differences between the two kinds of ionizing radiation (Lundqvist et al., 1968; Lundqvist and Lundqvist, 1988b; Lundqvist, 1992). In summary, the wealth of alleles distributed on a large number of cer loci has provided important insights into the mutation process. These insights into the mutation process combined with knowledge of the localization of the different genes in the genome will add to our understanding of the mechanisms of mutagenesis and the organization of the eukaryotic genome (Lundqvist and Lundqvist, 1988b; Lundqvist, 1992).

The frequency distribution of the number of mutations observed at cer loci for different mutagen treatments is shown in Fig. 2.6. The most frequently observed mutations are at the cer-c, cer-q and cer-u loci, all affecting the wax coating on spike and leaf sheath. Most mutations in the cer-c, cer-q and cer-u loci were obtained after treatment with chemicals. They are all located in the short arm of chromosome 2H, in a very closely linked gene cluster cer-cqu. The locus cer-b affecting wax on the spike and leaf sheath was lacking mutations after treatments with sodium azide; for the locus cer-i affecting wax coating on the spike, the greatest number of mutations was caused by densely ionizing radiation; while cer-j (affecting wax coating on leaf blade) had a high mutation frequency for treatment with chemicals, but no mutant isolated with neutrons. These results emphasize the variability in the frequency of mutations with respect to the mutagen, the gene or the combination of these two (Lundqvist and Wettstein, 1962; Lundqvist et al., 1968). Also 13 multiple cer-cqu mutants (seven triple and six double) were found, a rather high frequency of 2.5%; 11 were with neutrons, one with γ-rays and one with ethyl methanesulfonate (Lundqvist et al., 1968; Wettstein-Knowles and Søgaard, 1980; Lundqvist, 1992; Lundqvist, unpublished).

Fig. 2.6. Percentage of each cer mutation at each locus. Values are normalized for the five main kinds of mutagens (sodium azide, organic chemicals, ethyleneimine, neutrons and X-and γ-rays). The letters at the bottom of the diagram are the locus designations; underneath, the number of alleles at each locus is given. The numbers at the extreme right are the two classes of pooled loci with 19 down to 4 and 3 down to 1 mutation, respectively, for different mutagens. The numbers on the y-axis are the percentages (after Lundqvist and Lundqvist, 1988b; Lundqvist, 1992).

Recently the cer-cqu gene cluster has been cloned (Schneider et al., 2016) and the following results were found. The Cer-c gene is a chalcone synthase-like polyketide synthase designated diketone synthase with two exons. Cer-q encodes a lipase/carboxyl transferase with a single exon and the Cer-u (AK373499) gene is a P450 hydroxylase with five exons. It is now definitively shown that there are three different genes involved (Schneider et al., 2016). The Eceriferum gene glf1 (cer-zh) when investigated in the mutant cer-zh.54 was identified as an elongase component, β-ketoacyl-CoA synthase (CER-ZH/HvKCS1) (Li et al., 2018).

3.4 Erectoides or dense spike mutants

The Erectoides (dense spike) mutants were the first viable mutants induced by irradiation and the most commonly induced morphological changes in ear density. They are characterized by compact, dense spikes, implying that the spike rachis internodes are shorter than in the mother strain (Fig. 2.7). They generally possess a very stiff and often short straw. The first uppermost internode of the culm is generally longer than in the mother cultivar and the basal ones are shorter. In all, about 1270 such Erectoides (ert) mutants have been isolated at Svalöf and studied intensively; all different cultivars as mentioned before were applied. Among 222 investigated mutants, 31 ert gene loci could be established. Most of the loci have distinct phenotypic characteristics, 30 are recessive and one is dominant; 21 loci have been localized and spread over the seven barley chromosomes. Differences in the mutation spectrum could be noticed: three of the Erectoides (ert) loci could be identified as mutagen specific, where more than 80%, 70% and 50%, respectively, of the alleles were induced by irradiation. The analysed mutant number is too small to state any gene preference. Most of the alleles of the ert-b locus have mutated in one special cultivar, ‘Gull’. Many of the Erectoides mutants are fully viable and very promising from a practical point of view, and their productive capacity became tested continuously. High-yielding Erectoides were formed by mutation in several gene loci. The most outstanding one is Erectoides 32 in locus ert-k, which was released as the new very stiff-straw and high-yielding cultivar ‘Pallas’ in 1958 and was specially grown in Denmark and Great Britain (Gustafsson, 1941; Hagberg et al., 1952, 1958; Hagberg, 1953; Fröjer et al., 1958; Persson and Hagberg, 1968).

Fig. 2.7. Examples of some Erectoides mutants at different loci.

Already in the 1920s, cultivars with very short rachis internodes and a pyramid shape were found, and for these the name zeocriton (little barley) has been historically applied. This name was also applied to induced mutants having a similar phenotype (Hayes and Harlan, 1920). The dominant Erectoides-r (Ert-r) gene has been shown to be allelic to the Zeocriton 1 (Zeo1) locus; the DNA sequencing showed that the Zeo1 mutants occur in a Hordeum vulgare APELATA2 (AP2)-like transcription factor, HvAP2 (Nair et al., 2010; Houston et al., 2013); the dense spike phenotypes are a consequence of a perturbed interaction between microRNA 172 (Hv-miR172) and its corresponding binding site on the mRNA from the HvAP2 gene, which acts early in spike development to regulate turnover of HvAP2 mRNA. The Zeo1 and Ert-r mutants occur in the last intron of HvAP2, the binding site of Hv-miR172, and prevent cleavage of the HvAP2 mRNA (Nair et al., 2010; Houston et al., 2013). Another gene, Erectoides-m (ert-m), was identified as an orthologue of ERECTA in Arabidopsis thaliana (Zakhrabekova et al., 2015). Several ert-m alleles also carry the anthocyanin-less 1 (ant1) gene, which cannot be separated from the ert-m alleles. Sequencing of HvERECTA in the barley ert-m mutants revealed severe DNA changes in 15 mutants, including full gene deletions in ert-m.40 and ert-m.64. Both deletions additionally cause anthocyanin deficiency associated with the closely linked anthocyanin-less 1 (ant1) locus (Zakhrabekova et al., 2015).

4 The Use of Mutants for Plant Breeding

Once the work with artificial induction of mutations began, it was evident that mutation programmes should be regularly included in breeding programmes of crop plants. The application of mutation research in plant breeding was the most important stimulus. It was shown already in the 1950s and 1960s that the work at Svalöf through the joint work with barley breeders and scientists can be used as an example of how mutation breeding can be employed in a crop improvement programme (Gustafsson, 1963). The main interest was focused on macro-mutations. Several characters such as earliness, straw stiffness, higher yields, semi-dwarfs, protein content and disease resistance are of major importance. Not only new direct mutants, but also the indirect use of induced mutations was applied. In the latter case, breeding work changed modifying systems by crossing mutants with various established cultivars and selecting the best recombinants homozygous for the mutations. In the Swedish programme, the use of macro-mutations has proved to be more successful than recurrent mutagenic treatments (Gustafsson et al., 1971).

A rather large number of mutant cultivars of two-row barley were registered as originals and 15 Swedish ones were commercially released. Two of these cultivars, ‘Pallas’, a stiff-straw, lodging-resistant and high-yielding Erectoides mutant, and ‘Mari’, an extremely early photo- and thermo-insensitive mutant, were produced directly by X-irradiation. All other cultivars derive from crosses and backcrosses with the X-ray induced mutants ‘Pallas’, ‘Sv 44/3’ and ‘Mari’ (Table 2.6). The series of cultivars obtained after crossing were found to be agriculturally suitable for different parts of Scandinavia and other parts of the world. The aim of this work was to demonstrate that original mutant materials can be used successfully in recombined breeding programmes and in the hands of skilful breeders. Additional methods ought to be used together, also today with many modern technologies, adding to the results of ordinary crossing and selection (Gustafsson, 1969, 1986).

Table 2.6. Induced barley mutants and their derivatives as released cultivars at Svalöf.

In conclusion, the words of Åke Gustafsson, the father of mutations, from his last paper (Gustafsson, 1986) are summarized as follows.

Useful mutations in barley include a wide range of economically important characters that influence morphological as well as physiological and biochemical properties and will be an important tool in plant breeding, even more when the chemistry of the gene has been studied more intensively. Genetic instruments of artificial selection will increase the power and capacity of the plant breeder. It seems rather strange that also today there is a certain negative attitude towards the use of mutations in plant breeding or in most experiments concerning the general evolutionary theory. Such negative ideas are often associated with the view that mutationists ignore the natural sources of genetic variability and oppose the breeding value of primitive biotype collections.

References

Bothmer, R. von et al. (1995) An ecogeographical study of the genus Hordeum. Systematic and Ecogeographic Studies on Crop Genepools 7 (2nd edn). IBPGR, Rome.

Bull, H. et al. (2017) Barley SIX-ROWED SPIKE3 encodes a putative Jumonji C-type H3K9me2/me3 demethylase that represses lateral spikelet fertility. Nature Communications 8, 936.

Comadran, J. et al. (2012) Natural variation in a homolog of Antirrhinum CENTRORADIALIS contributed to spring growth habit and environmental adaptation in cultivated barley. Nature Genetics 44, 1388–1392.

Dormling, I. and Gustafsson, Å. (1969) Phytotron cultivation of early barley mutants. Theoretical and Applied Genetics 39, 51–61.

Dormling, I. et al. (1966) Phytotron cultivation of Svalöf’s Bonus barley and its mutant Svalöf’s Mari. Hereditas 56, 221–237.

Druka, A. et al. (2011) Genetic dissection of barley morphology and development. Plant Physiology 155, 617–627.

Ehrenberg, L. et al. (1952) Effects of ionizing radiation in barley. Arkiv Botanik, Ser. 2.1 No. 17, 557–568.

Ehrenberg, L. et al. (1956) Chemically induced mutation and sterility in barley. Acta Chemica Scandinavica 10, 492–494.

Ehrenberg, L. et al. (1959) The mutagenic effects of ionizing radiation and reactive ethylene derivates in barley. Hereditas 45, 351–368.

Ehrenberg, L. et al. (1961) Viable mutants induced in barley by ionizing radiations and chemical mutagens. Hereditas 47, 243–282.

Esse, G.W. van et al. (2017) Six-rowed spike3 (VRS3) is a histone demethylase that controls lateral spikelet development in barley. Plant Physiology 174, 2397–2408.

Franckowiak, J.D. and Lundqvist, U. (2001) Rules for nomenclature and gene symbolisation in barley. Barley Genetics Newsletter 31, 76.

Fröjer, K. et al. (1958) Pallas barley, a variety produced by ionizing radiation: its significance for plant breeding and evolution. In: United Nations Report 2nd Conf. on the peaceful uses of atomic energy in Geneva, August1958.

Gustafsson, Å. (1938) Studies on the genetic basis of chlorophyll formation and the mechanism of induced mutating. Hereditas 24, 33–93.

Gustafsson, Å. (1940) The mutation system of the chlorophyll apparatus. Lunds Univ. Årsskrift, N.F. Avd. 2. Bd. 36, 1–40.

Gustafsson, Å. (1941) Mutation experiments in barley. Hereditas 27, 225–242.

Gustafsson, Å. (1942) Mutationsforschung und Züchtung. Der Züchter 14, 57–64.

Gustafsson, Å. (1947) Mutations in agricultural plants. Hereditas 33, 1–100.

Gustafsson, Å. (1954) Swedish mutation work in plants: background and present organization. Acta Agriculturae Scandinavica IV 3, 361–364.

Gustafsson, Å. (1963) Productive mutations induced in barley by ionizing radiations and chemical mutagens. Hereditas 50, 211–263.

Gustafsson, Å. (1969) Positive Mutationen und ihre Verwendung in der Züchtung hochleistender Gerstensorten. Bericht über die Arbeitstagung der Arbeitsgemeinschaft der Saatzuchtleiter in Gumpenstein, November 1969, pp. 63–88.

Gustafsson, Å. (1986) Mutation and gene recombination – principal tools in plant breeding. In: Olsson, G. (ed.) Svalöf 1886–1986.Research and Results in Plant Breeding. LTs förlag, Stockholm, pp. 76–84.

Gustafsson, Å and Lundqvist, U. (1976) Controlled environment and short-day tolerance in barley mutants. In: Induced Mutations in