**3. Lack of formal genetic improvement**

for adaptation to stressful environments such as water stress, salinity and high temperatures. A systematic landrace evaluation may define patterns of diversity, which will facilitate identifying alleles for enhancing yield and abiotic stress adaptation, thus raising the productivity and stability of staple crops in vulnerable environments. It can also be defined as a traditional variety with a high capacity to tolerate biotic and abiotic stresses, resulting in highyield stability and an intermediate yield level under a low-input agricultural system [3]. A landrace differs from a variety that has been selectively improved by breeders for particular

Landraces are important genotypes for crop breeding owing to their high potential to adapt to specific environmental conditions and the large source of genetic variability that they provide [4]. Landraces are generally less productive than commercial cultivars, although in recent years, they have become important as sources of genetic variability in the search for genes for tolerance or resistance to biotic and abiotic factors of interest in agriculture [5]. The genetic diversity observed across landraces is the most important part of maize biodiversity, and local races represent an important fraction of the genetic variability exhibited by this genus. However, few agronomic and genetic data exist for such collections, and this scarcity has limited the use, management and conservation of this germplasm. In addition, a few improved genotypes with narrower genetic variability are quickly replacing maize

Zeven opined that landraces have played a fundamental role in the history of crops worldwide, in crop improvement and agricultural production, and they have been in existence since the origins of agriculture itself. During this time they have been subject to genetic modification through abiotic, biotic and human interactions. For centuries, crop landraces were the principal focus for agricultural production [7]. Farmers sowing, harvesting and saving a proportion of seed for subsequent sowing over millennia have enriched the genetic pool of crops by promoting intraspecific diversity [8]. This cycle remained current until the dawn of formal plant breeding and the generation of generally higher-yielding cultivars that subsequently

The origin of landraces encompasses both the temporal and spatial components of where landraces were first developed. They (landraces) have a relatively long history, significantly more than the ephemeral lifespan of modern cultivars. Many authors suggest that landraces have been growing 'since time immemorial' [11], 'over long periods of time' [9], 'over hundreds even a thousand years' [12], 'for many years even centuries' [13], 'for generations' [14], 'for many centuries' and 'over a period of time' [15]. Nevertheless, few are explicit about the amount of time a landrace must be grown to be considered a landrace. However, Louette [16] indicated for maize that the period of time must be 'for at least one farmer generation (i.e. more than 30 years)', while Astley referred to vegetable landraces being grown for '50–70 or

characteristics.

4 Rediscovery of Landraces as a Resource for the Future

landraces [6].

even 100 years'.

replaced many traditional landraces [7, 9, 10].

**2. Historical background of landrace origin**

Landrace production is associated with 'no human selection' [11] and 'it was naturally developed' [23]; thus, landraces have been developed as a result of time and natural selection alone. Other authors suggest that human selection has occurred but in the form of unconscious selection, and others suggest that a certain degree of consciousness is involved in the selection process, 'without or with only little mass selection' [23], 'subject to some deliberate selection' [24], 'artificial selection (probably largely of an unconscious nature)' [17] and 'breeding or selection … either deliberately or not' [14]. Where conscious human selection has been recognized as being significant in landrace development, it has nevertheless been distinguished that is applied to modern cultivars [7, 12] with qualifications such as 'more resistant to pests and diseases, have more yield stability' [25], 'grown in traditional farming systems' [7, 13], 'cultivated in low-input cultivation' [8], 'in a number of traits which together appear to form an adaptive complex' [3] and 'on a low selection pressure'.

It is generally accepted that farmers, gardeners and growers select and develop landraces [12–15, 17, 26], while formal plant breeders select and develop cultivars (**Figure 1**). However, even this division is not as clear as it first may appear if other considerations are included. Zeven [27] explained that 'continuous selection by some farmers for plants with desired characters is similar to the later proposed scientific selection within landraces to select by seekers for the best plants'. Examples of these are shown in vegetables that present special traits such as enormous size, developed by growers in the UK.

The situation concerning the involvement of landraces in participatory plant breeding is interesting, as Maxted [28] noted that care should be taken to ensure the security of the locally adapted genetic diversity or the former landrace could no longer be regarded as a landrace. Here, the decision over whether the former landrace may still be regarded as a landrace as described by Almekinders and Elings [29] depends on the degree of breeding and the quantity of external germplasm introgressed with the original landrace; the more of either the less the entity could be regarded as a landrace. Certainly, this would be the case for participatory varietal selection programs where external germplasm is introduced into an area and suitable material is selected by local farmers; even if the new germplasm is managed by the farmer in a manner usually associated with traditional farming and landrace maintenance, the use of the term landrace would be inappropriate. Yet, another consideration is understood by the term 'modern' crop improvement.

Simmonds [30] and Allard [31] further explained that modern professional crop improvement is based on the Darwinian theory of evolution through selection and the genetic mechanisms of evolution developed by Mendel, Johannsen, Nilsson-Ehle, East and others. Frankel and Bennett [9] used as a reference point the 19th century when conscious, individual plant selection commenced. Jarman and Leggett considered that 'modern' crop improvement started when formal breeding programs were initiated, in the UK, for example, in the 1920s. However, the fact that the history of crop improvement is different for each crop is also an important element to be considered [3]. Combining these considerations, formal crop improvement is understood as the application of genetic principles and practices to the development of cultivars by both classic breeding techniques (selection and hybridization) as well as more recent technologies (biotechnology, molecular biology, transgenics) within a crop improvement program. Virchow [32] when defining the characteristics of a landrace included the fact that landraces are not registered in official seed lists, but in the UK, several entities generally regarded as landraces, such as Kent Wild White Clover, are included on the National List and are regarded as landraces because they result from farmers' selection over millennia.

improvement programs, in contrast with modern cultivars which have resulted directly from

Landraces and Crop Genetic Improvement http://dx.doi.org/10.5772/intechopen.75944 7

Despite this improved clarification, there remains confusion as regards the effect of crop evolution on landraces. Crop evolution is not a linear process, and there are different points of view of the position occupied by landraces in relation to their wild relatives, on the one hand, and cultivars, on the other. Some authors such as Marchenay [34] suggested that some landraces exist on the borders of cultivation, not having been fully domesticated and might be better considered as ecotypes. Other authors raised the issue that some landraces have crossed freely with their wild relatives over millennia [24, 35] and as a result possess rudimentary characters or 'wild relative traits' not found in cultivars because of their more ephemeral existence. While

these programs (**Figure 1**).

**Figure 1.** Different opinions about the types of landrace selection [36].

In fact, it is argued that inclusion of landraces on the UK National List is likely to promote their cultivation and thus conservation [33]. Landraces may therefore be more easily defined as being crop varieties which do not result in the first instance at least from formal crop

#### Landraces and Crop Genetic Improvement http://dx.doi.org/10.5772/intechopen.75944 7

**Figure 1.** Different opinions about the types of landrace selection [36].

'grown in traditional farming systems' [7, 13], 'cultivated in low-input cultivation' [8], 'in a number of traits which together appear to form an adaptive complex' [3] and 'on a low selection

It is generally accepted that farmers, gardeners and growers select and develop landraces [12–15, 17, 26], while formal plant breeders select and develop cultivars (**Figure 1**). However, even this division is not as clear as it first may appear if other considerations are included. Zeven [27] explained that 'continuous selection by some farmers for plants with desired characters is similar to the later proposed scientific selection within landraces to select by seekers for the best plants'. Examples of these are shown in vegetables that present special traits such

The situation concerning the involvement of landraces in participatory plant breeding is interesting, as Maxted [28] noted that care should be taken to ensure the security of the locally adapted genetic diversity or the former landrace could no longer be regarded as a landrace. Here, the decision over whether the former landrace may still be regarded as a landrace as described by Almekinders and Elings [29] depends on the degree of breeding and the quantity of external germplasm introgressed with the original landrace; the more of either the less the entity could be regarded as a landrace. Certainly, this would be the case for participatory varietal selection programs where external germplasm is introduced into an area and suitable material is selected by local farmers; even if the new germplasm is managed by the farmer in a manner usually associated with traditional farming and landrace maintenance, the use of the term landrace would be inappropriate. Yet, another consideration is understood by the term 'modern' crop improvement.

Simmonds [30] and Allard [31] further explained that modern professional crop improvement is based on the Darwinian theory of evolution through selection and the genetic mechanisms of evolution developed by Mendel, Johannsen, Nilsson-Ehle, East and others. Frankel and Bennett [9] used as a reference point the 19th century when conscious, individual plant selection commenced. Jarman and Leggett considered that 'modern' crop improvement started when formal breeding programs were initiated, in the UK, for example, in the 1920s. However, the fact that the history of crop improvement is different for each crop is also an important element to be considered [3]. Combining these considerations, formal crop improvement is understood as the application of genetic principles and practices to the development of cultivars by both classic breeding techniques (selection and hybridization) as well as more recent technologies (biotechnology, molecular biology, transgenics) within a crop improvement program. Virchow [32] when defining the characteristics of a landrace included the fact that landraces are not registered in official seed lists, but in the UK, several entities generally regarded as landraces, such as Kent Wild White Clover, are included on the National List and

are regarded as landraces because they result from farmers' selection over millennia.

In fact, it is argued that inclusion of landraces on the UK National List is likely to promote their cultivation and thus conservation [33]. Landraces may therefore be more easily defined as being crop varieties which do not result in the first instance at least from formal crop

as enormous size, developed by growers in the UK.

6 Rediscovery of Landraces as a Resource for the Future

pressure'.

improvement programs, in contrast with modern cultivars which have resulted directly from these programs (**Figure 1**).

Despite this improved clarification, there remains confusion as regards the effect of crop evolution on landraces. Crop evolution is not a linear process, and there are different points of view of the position occupied by landraces in relation to their wild relatives, on the one hand, and cultivars, on the other. Some authors such as Marchenay [34] suggested that some landraces exist on the borders of cultivation, not having been fully domesticated and might be better considered as ecotypes. Other authors raised the issue that some landraces have crossed freely with their wild relatives over millennia [24, 35] and as a result possess rudimentary characters or 'wild relative traits' not found in cultivars because of their more ephemeral existence. While

**4. Recognizable identity**

populations with a similar trait'.

**5. High genetic diversity**

changes.

consistently used to define the identity of a specific landrace.

Landrace must be intrinsically highly genetically diverse and recognized as a distinct entity via common-shared traits. These traits will allow the distinction of one landrace from another or from modern cultivars for the same crop. They will sometimes give rise to landrace names, but at other times, names may be determined by other factors such as use or origin. Therefore, landraces 'are each identifiable and usually have local names' [7], 'are recognized morphologically' [14], 'have a local name,' 'are a farmer selection based on local characteristics (specific use, local market, horticultural practices and locally adapted)' and 'are heterogeneous

Landraces and Crop Genetic Improvement http://dx.doi.org/10.5772/intechopen.75944 9

However, this characteristic may be difficult to be applied universally as landraces identified on the basis of common names can be misleading because of non-associated synonyms and homonyms. Many disparate landraces may be named after their early flowering capability or seed color, for example. A landrace may be recognized by different names in different countries or communities [36], or conversely quite different landraces can be designated with the same name [14]. These factors contribute to one of the main problems associated with landraces, namely, their consistent identification and the determination of which traits can be

The characteristics of landraces in relation to the magnitude of allelic and genetic diversity in contrast to cultivars are considered to be significantly more genetically diverse [37]. Thus, a landrace is a 'highly variable population in appearance' [7], 'highly diverse populations and mixtures of genotypes' [38], 'genetically heterogeneous' [13], 'not genetically uniform and containing high levels of diversity' [14], 'local diverse crop varieties' [26], 'heterogeneous crop populations' [39] and 'materials with variable levels of heterogeneity'. Frankel and Soule [40] indicated that the genetic diversity of landraces has two dimensions: between sites/populations and within sites/populations. The former is generated by heterogeneity in space and reproductive isolation, while the latter is generated by heterogeneity in time associated with both shortterm variations between seasons and by longer-term climatic, biological and socio-economic

Some authors have used the term 'meta-population' when referring to the diversity structure of a landrace. As such, a landrace constitutes a group of farmers' seed lots that are highly diverse both between and within themselves. In contrast however, Sanchez [41] when evaluating the genetic diversity of maize landraces of Mexico found that some landraces had very low levels of genetic diversity, and it was suggested that comparatively low diversity may be more associated with selfing crops. Bere barley, one of the oldest cereal varieties in Europe, is 'surprisingly homozygous', possibly because it has been maintained in isolation in marginal lands since the sixteenth century [42]. A similar picture is provided by Tibetan barley landraces

**Figure 2.** Types of landraces of maize [27].

others believe that landraces can even be selected from cultivars [18, 19], terms such as creolization or rustication are applied, and 'in the absence of traditional and formal maintenance breeding, any improved landrace (cultivar), including a hybrid variety, will regress with time into a landrace' [27]; 'a cultivar that has been growing under a low selection pressure for specific traits but not uniformity for a long time could be considered a landrace' (**Figure 2**).
