A review on correlation, heritability and selection in silkworm breeding

Abstract

Silkworm, Bombyx mori, is one of the most significant insects in silk production as well as in biological studies. The aim of silkworm breeding is to gradually improve economically valuable traits to increase the profits of the sericulture industry. This article reviews the necessity and achievements of research based on genetics and breeding. Some genetic principles in silkworms, along with the goals and outcomes of breeding are reviewed briefly. Selection under different environmental conditions, calculation of breeding values, quantitative and qualitative traits, consanguinity and heritability are also discussed.

Keywords:

• Bombyx mori
• genetic improvement
• breeding value
• production potential
• silkworm

1. Introduction

While rearing any type of animal, several issues should be considered. The first issue is the manner of animal rearing in which nutrition is one of the important factors. Hygiene is the second issue which may result in increased damage and decrease in production capability. The third issue is breeding which heightens the production capability of an animal. If the herd is not selected for a specific goal and functional improvement from one generation to the next, it will not functionally develop. Animal breeding is the science and art of effective gene stabilization in economically important animal production (Tahmoorespour 2007).

The native mass of each species in a country is considered as national capital and an essential and key product. Hence, their protection and multiplication are worthwhile. These animals have survived and multiplied generation after generation for thousands of years, passing many obstacles and disagreeable and unfavourable conditions (Sheikhlou et al. 2008). Now, the survival of the sericulture industry depends on increasing both the quantitative and qualitative efficiency of silkworm production and also improving the output of products from this industry.

Increases in the efficiency of silk production (the basic product of the sericulture industry) requires the employment of scientific methods of breeding, provision of standard conditions for adequate nutrition and maintenance of environmental temperature, adequate humidity, light, sanitation, etc. It also involves the use of genetics and breeding for accumulation of desirable genetic factors in commercial species of silkworm (Javanshir 1996). Silk as the primary substance is used in the production and manufacture of carpet, cloth, shirts, scarves, wall hangings and home decoration, and also in the construction of parachute fabric, tyre lining, artificial blood vessels, surgery thread, military instruments and electric insulator substances, etc. Moreover, the silkworm is used in fundamental research such as genomic studies and in the provision and acquisition of connection designs based on different morphologic (He et al. 2001) and molecular (Mirhosseini et al. 2012) markers.

The Berry silkworm (Bombyx mori) is the most economically important species for natural silk production in the sericulture industry (Qeyasi 1995). More than 95% of natural silk all over the world is obtained from this insect (Collard et al. 2005). These days, conservation of genetic resources is one of the most important goals of breeding science (Tavakolian 2000). The aim of this review is to summarize the latest achievements and research results about correlation, heritability and selection in silkworm breeding.

2. Genetic principles in animal science: a brief review

Humans have long been trying to comprehend the mechanisms of inheritance. The principals of this modern science go back to Gregor Mendel (1822–1844) who related inheritance of traits to transmission of an inheritance unit among generations. The inheritance unit, which is now called a gene, must have two characteristics: (1) to be able to transfer from one generation to another in a way such that each generation contains one copy of it and (2) to include information about construction, function and other biological processes for these generations. The recent developments in genetics or modern genetics can be divided into three main categories (Yazdi Samadi and Sayed Tabatabaei 2008): (1) Mendelianor classical genetics which studies the transmission of traits from one generation to another, (2) molecular genetics which discusses chemical construction of genes and investigates at the molecular level, and (3) population genetics which studies variety in the population itself or among different societies.

The productive potential of every creature depends on genetics, environmental factors and the interaction between these two factors. The genetics of all creatures plays an essential role in their productive potential. In recent years, studies in animal science have included the exploration of genetic potential and animal breeding of economic insects such as the honeybee and silkworm. Accordingly, many studies have been accomplished about genetics and silkworm breeding in Iran and other countries of the world (Salehinezhad 2010b; Neshagaran Hemmatabadi 2011). Genetic variety plays an important role in the advancement of animal and plant industries for two reasons:

1. The capability for identifying genetic variety which is considerable in breeding programmes and population genetics analysis; and

2. The estimation of variety rates in genotypes which is essential for the calculation of increase in genetic potential in breeding programmes and important for the adjustment of cross-breeding programmes (Nagaraju & Singh 1997).

The increase of utility due to genetic improvement of domesticated animals has been a focus for breeding experts for a long time. The amount of breeding success in production traits for the improvement of animals is connected with the inheritance dimension of observed variations among animals, identification of individuals possessing eligible genes and their selection as the parents for the next generation (Ahmadi Shahrakht et al. 2012).

3. Silkworm genetics: a brief review

The silkworm is important not only for silk production but also for biological studies. After Drosophila melanogaster, the bulk of studies on insect genetics relates to the silkworm. Almost 4310 silkworm strains (including geographical strains, consanguineous lines and mutants) have been identified. Of these, about 2000 strains belong to the B. mori L silkworm. In addition, about 230 gene and locus patterns have been recognized (Yoshido et al. 2005; Zanatta et al. 2009).

Because of the deep-rooted history of silkworm nurturing for economic benefits, silkworm genetics has been an important topic among scientists. This significance has resulted in the collection of, study of and conservation of silkworm genetics for scientific and economic utilization (Doria et al. 1992). Additionally, the genetics of successful traits of silkworms have been well investigated in scientific sources which have put the emphasis on qualitative traits. The appearance of these traits for one or several genes can be identified easily. So, on the basis of these genetic traits, we can anticipate possible variation intervals of a specific trait while regenerating and improving another trait (Seidavi & Bizhannia 2008a).

The number of chromosomes in the domesticated silkworm (B. mori) is 28, while in the wild silkworm (Bombyx mandarina) it is 31. Moreover, the wild silkworm contains 27 haploid chromosomes (Nematnezhad 1998). More than 450 morphological and biochemical characteristics have been recorded for the silkworm. Of these, 300 characteristics have been determined on 27 groups of chromosomes (Cristina et al. 2007). The reasons are: (1) the mature insect has a short longevity and the nymph is latent in the cocoon so the chance of exposure is too small; and (2) larvae are killed after the cocoon spinning phase in order to procure silk and only a few of them are used for producing eggs (Shabdini Pashaki 2010).

Several investigations have been carried out on the genetics, nutrition and biochemistry of silkworms (Seidavi et al. 2008b, 2009; Seidavi 2009, 2010a, 2010b, 2011a, 2011b). In 1928, Kavagushi had crossed domesticated silkworms (B. mori) with wild silkworms (B. mandarina) and observed that one of the 27 chromosomes of the wild silkworm paired with two chromosomes of the domesticated silkworm and resulted in a viable cross-breed. He concluded that one of the B. mandarina chromosomes had fractured and changed into two parts and thereupon, giving rise to the domesticated silkworm. Meiosis immediately begins after fecundation. When the embryo formation is complete, the larvae break the egg shells and emerge. The newborn larvae are black and grey and covered with hair. For this reason, it is called Hair worm or Ant worm (Neshagaran Hassanabadi 2006).

Numerous traits are significant in sericulture and the importance of these traits varies in different sections of the industry. Silkworm egg producers aim to possess lines with high breeding potency, while cocoon producers endeavour to improve production, cocoon shell per cent and resistance against illness (Sing et al. 1998). At present in Iran, silkworm egg producers use individual cross-breds which are the result of crosses among Japanese lines 31, 103, 107, 151, 153 and Chinese lines 32, 104, 110, 154, in order to produce commercial silkworm eggs (Mirhosseini et al. 2009). Variation in the breeding programmes results in the construction of a silkworm population with a firm base. Moreover, evaluation of genes helps in distinguishing lines via specific characteristics such as the length of fibre, denier, resistance against tension, resistance against illness, etc. (Li et al. 2001). Rao (1997) reported that heritability increases because of an increase in the genetic improvement of some traits such as individual shell weight for bivoltine lines and cocoon weight, shell weight and fibre length for multivoltine species. High heritability and medium genetic improvements for larvae weight and individual cocoon weight in multivoltine species have been attained (Sen et al. 1995).

4. Silkworm breeding

4.1. Aims and achievements

The aim of silkworm breeding is the genetic improvement of traits to increase profitability of the sericulture industry (Groen 1990; Sing et al. 1998; Ghanipoor et al. 2006; Talebi & Subramanya 2009; Mubasher et al. 2010; Salehinezhad 2010a; Seidavi 2010a, 2010b, 2010c). They aim to do this by gradually improving economic traits of the silkworm to provide benefits and increase the profits of other parts of the sericulture industry (Mirhosseini et al. 2005). It has been estimated that about 3000 varieties of silkworm have been trial led under various breeding programmes all over the world (Nagaraju 2002). The best features can be incorporated through breeding and the best silkworms as regarding purification and alteration for improvement can be determined (Seidavi 2010a). Silkworm breeding has resulted in higher cocoon production, a rapid increase of raw cocoon yield and better quality (Seidavi and Bizhannia 2008a; Neshagaran Hemmatabadi et al. 2011b).

The aims of silkworm breeding are to achieve better function in: (1) silkworm egg production, (2) cocoon and raw silk production, (3) continuance of production in variable conditions, (4) compatibility and production capability in new locations and (5) cocoon and raw silk quality. In addition, we can refer to other traits such as the ability of feeding from different nutrition sources; the use of artificial rations; the increased resistance against illness, pests and extreme environmental conditions; physical characteristics related to size and other qualitative particularities of silk thread; development of some behavioral traits in larvae based on new methods and mechanization of rearing and the picking up of cocoon (Seidavi & Bizhannia 2008a). Overall, the increase of cocoon output by the increase of fibres, and the increase of fertility as well as the increase of viable birth results in an efficiency improvement in silkworm breeding programmes (Chatterjee 1993; Strrunnikov 1995). Selection is one of the best approaches in any silkworm breeding programme. Individual selection in parental lines can improve functional results, but this improvement is affected by environmental factors especially those related to the season of the year. Seidavi (2010a) found that season had a considerable effect on silkworm function with more selection response in spring than in autumn. In the experiments of Nadarajan and Gunasekran (2008), breeding pressure in three lines was not significant, except for Mysore and Nistari.

4.2. Expected breeding value

Domesticated species are nowadays considered as national capital because of their natural selection for so many years and their endurance of different environmental factors. In addition, they are referred to as strategic production because they increase efficiency due to lower expenses. Genetic reserves and genetic variety are important goals of breeding (Davis & Denise 1998; Dekkers 1999; Davis et al. 2002). The calculation of additive genetic pure value used in animals includes specific statistical methods such as Best Linear Unbiased Prediction (BLUP). Adequate selection strategy requires that important factors should be considered such as the type and the number of important economic traits, genetic relation among the traits, heritability of traits and the degree of genetic and economic response to selection (Mirhosseini & Seidavi 2008a). For the effects of selection type on economic value and selection indicator in different animals, Bizhannia et al. (2006) have studied the effect of parental selection on genetic parameters in general and specific combination ability and heterosis for pure silkworm lines from economic, breeding and heterosis aspects. They showed that selection increases production and reproduction efficiency and also that heterosis selection and general and specific combination ability increase reproduction indicators (Mirhosseini & Seidavi 2008b).

Additive genes influence many important traits such as the growth amount, milk production, shape, quantity and quality of carcass in farm animals. There are so many polygenic traits, which are chiefly distinguished by additive genes, have average to high heritability and are so little influenced by cross-breeding and relative rearing (Amanlou 2005). The success of a breeding programme is dependent on the amount of genetic variety in a population. Genetic variety is the main principle for animal breeders and rarity or lack of this variety decreases the power of genetic selection (Barker 1994). Moreover, an increase in heritability causes the increase of additive genetic activity and the improvement of response to selection (Petkov & Nguyenvan 1987).

In many localities of the world, the comparison between the function of hybrid species with that of domesticated species is done based on weak projects which mostly yield false results (FAO 1998). With silkworms, the mean of cocoon weight breeding value in various pure lines, and selection and non-selection act among the groups to demonstrate that significant variance exists in the genetic values among pure lines indicating random genetic direction which causes changes of gene structure in different silkworm populations. Genetic function in a selected population decreases in inadequate environmental conditions arising from an increase in sensitivity to inadequate environmental factors, and the increase of illness and death caused by the selection based on production factors (Seidavi 2010a).

The silkworm is one of the most important and the most profitable commercial insects in many countries for the purpose of raw silk production and as a source for biological substances. With the progress of genomic technology, access to information about silkworms such as genome sequencing and genetic markers has accelerated (Kim et al. 2010). Various genotypes have shown that there are many differences in quantitative and qualitative traits which have an important part in silkworm efficiency (Nagaraju & Singh 1997). In the mulberry silkworm, 21 characteristics have been distinguished which influence silk production quantitatively and qualitatively (Chatterjee 1993; Pal & Moorthy 2011). Some characteristics are inherited and some others arise from environmental factors. The environment plays an essential part in economic traits which are quantitatively inherited (Singh et al. 1994; Rao et al. 2004). For example, in a study about the effect of environmental factors on silkworm efficiency, Gangwar (2012) concluded that varieties being studied showed the best compatibility in spring and autumn while unsuitable functions were observed in the hot summer season and in the rearing season. In addition, Morohoshi (1969) has demonstrated that seasonal function in bivoltine strains depends on the silkworm technology in tropical temperature conditions. Various genotypes among the lines show various responses to changing environmental factors as a result of variable gene frequency in different areas and in the arrangement of varieties. Seidavi et al. (2004b) and Govindan et al. (1991) reported that the cocoon weight trait and the cocoon shell weight trait were influenced by additive genes and dominant genes; and also that additive genetic variance is more influential on traits than dominant genetic variance.

4.3. Considered quantitative traits in the silkworm breeding

Sericulture is the science of silkworm rearing for efficient raw silk production which is then exploited for silk fibre production (Krishnaswami et al. 1973). Silkworm in different geographical areas should have a considerable interest in breeding programmes (Jolly et al. 1989). Economic coefficients of traits, which are defined as the contribution of those traits in the improvement of economic function of production systems, indicate the quantitative aims of breeding (Groen 1989; Ghanipour et al. 2006). Japanese and Chinese breeders have made noticeable advancements through the improvement of important economic traits in silkworm (Harada 1961; Gamo 1976; Mano et al. 1991; Chen et al. 1994).

The silkworm has many varieties with respect to geographical areas (rearing area) and breeding lines, which differ from each other in many quantitative traits (Reddy et al. 1999). The most important economic traits connected with silk production include the weight of one cocoon (g), the weight of one shell (g), the per cent of shell, cocoon efficiency (g) for every 10,000 larvae, fertility (per cent), the cost of pupa establishment (per cent), the length of larvae period (hour), the average length of fibre (m), unbroken fibre length (m), denier (g), spinning capability (per cent) and raw silk (per cent; Kumaresan et al. 2003). Mirhosseini et al. (2012) consider that important economic traits in silkworms such as cocoon weight, the per cent of cocoon shell, the length and the thickness of silk thread, fertility and resistance against bacillus virus are accounted as quantitative traits, and are controlled by numerous genes each small effects and environmental conditions. Moreover, Kang et al. (2001) have stated that among economic characteristics of silkworm the population number, weight of population, cocoon shell weight, and the per cent of cocoon shell are significant influential factors on the increase of cocoon efficiency.

In silkworms, quantitative traits are controlled by numerous genes and environmental factors (Reza & Rahman 2005; Zhao et al. 2007; Ahsan & Ramian 2008). Therefore, the production potential of silkworm needs to be improved through breeding programmes. The aim of silkworm breeding is to access a higher function of traits such as egg efficiency, raw cocoon silk efficiency, cocoon stability and production following development of other new parts (Singh & Kumar 2010). The major effect of breeding is the decrease of the variation of the population; as a result, gene frequency of the population is fixed, but its genotype frequency changes more in homozygotes and less in heterozygotes (Kumar 2005). Accordingly, every change in the mean of population caused by breeding is connected with differences of genotype value among heterozygotes and homozygotes (Singh & Kumar 2010). Iftekher et al. (2005) have investigated the amount of heterosis variety for different indicators through maximum heterosis for cocoon shell weight, cocoon weight and silk fibre thickness. The accounted advantages of the breeding programme are accurate and variable only when the direct necessity of all the related reserves are considered (Cunningham 2003).

If a metric trait is influenced by a gene locus, to a great extent this effect is related to the amount of a gene product. Regulator genes, which control the amount of gene product, may be the main source of changes in metric traits (McDonald & Ayala 1978; Hedrick & McDonald 1980). In the studies of Kumar & Kumar (2011), estimation of cocoon weight was about 1.476–1.890 g; and so, increase of record (1.890 g) for line V-2 and decrease of record for line S-36 (1.476 g) were obtained. The limit of cocoon shell weight was between 0.241 and 0.351 r, that is the highest record was for the V-1 line (0.351) and the lowest record was for the S-36 line (about 0.241). Also, the per cent of cocoon fibre was about 14.79–19.58%; and so, the highest record was in line V-2 (19.58%) and the lowest record was in line S-36 (14.79%). The traditional method of breeding includes hybridization which is done at the most suitable times of the year. Now, some traits in silkworm such as the form of cocoon, the colour of cocoon, the length of silk fibre, larvae survival and some other one are used in different varieties (Reddy et al. 1999).

4.4. Considered qualitative traits in the silkworm breeding

The genetic variety in any species of domesticated animal is a source for making necessary changes in phenotypic characteristics of the population. These characteristics can be generally divided into two categories: production traits (quantitative and qualitative) and competence traits (compatibility, adjustment, fertility and resistance against illness; Gharahveysi et al. 2010). The discussion of economic traits is not the only topic in sericulture as there are other important traits (Nagaraju 2002). From considered qualitative traits in silkworm, we can refer to the following traits: egg colour, egg chorionic colour, the colour of body in the last larvae period, the form of body in the last larvae period, the signs of the cocoon, the size of the cocoon and the tissue of raw cocoons (Cristina et al. 2007). Seidavi and Bizhannia (2008a) consider that the most important qualitative traits for silkworm breeding experts include the colour of egg, the signs on the larvae body, the colour of the blood, the colour of the cocoon and the form of the cocoon. Another important qualitative trait is the colour of the eyes of the silkworm butterfly which has a close connection with the colour of the silkworm egg.

It is essential to know that multiple genes and alleles are responsible for the separation of cocoon colour (Tazima 1964). Chiefly, the colours of cocoons are yellow, golden yellow, pink, pale green or white. Lim (1955) found that yellow and pink colours are because of the existence of carotinoids which is the result of mulberry leaf consumption containing carotenes and xanthophylls. Previous studies demonstrated that with the cross between pure multivoltine Mysore varieties having a green cocoon and the bivoltine hybrid cocoon (NB7*NB4D2), in the first species all the cocoons were green and in the second species cocoon colours were separately green, pale yellow, pale white and white. Accordingly, it is supposed that one or several genes are responsible for generating of cocoon colour by several alleles in various domesticated species (Goldsmith et al. 2005; Kumar et al. 2009). Farooq et al. (2002) declared the positive effects of hybrid conjugation for cocoon traits. Usually, some traits in silkworm such as cocoon form, cocoon colour, the length of silk fibre, larvae survival and other traits are investigated in different varieties and in the act of parental selection (Kumaresan et al. 2000; Rao et al. 2006; Moorthy et al. 2007; Hussain et al. 2010).

4.5. Consanguinity

Many studies are available on commercial matings between two or more of prior forms (species or lines) from different geographical sources. One of the most important methods is selection and the use of heterozygosity increase of the silkworm in the first species of hybrid. However, the increase of new cocoon production in relation to the increase of silk production is not always the same. Because silk production is one of the greatest economic goals in cloth-weaving factories, many scientists pay specific attention to the evaluation of silk production ability in the first hybrid species in heterosis (Petkov et al. 2007a). Consanguinity is harmful for the silkworm as well as the cocoon production. By suitable matings, decrease of cross-bred superiority and the reduction of fertility in consanguine species can be bred. Silkworm breeding experts purpose fully use cross-bred superiority as an important method in genetic breeding of characteristics related to silken cocoon production (Seidavi & Bizhannia 2008a).

Artificial selection is widely applied in breeding programmes of important economic insects. Selection increases hemozygote frequency and homozygosity influence (Whitlock 2002). Avendano et al. (2004) have stated that a breeding programme with an adequate consanguinity limitation can increase genetic selection efficiency from 20% to 25% in a specific level of consanguinity rate. Meuwissen (1997) have investigated circumstances of maximizing genetic improvement with limited consanguinity. The direct effect of this method is that with limitation of consanguinity rate, the influential size of a fixed population is conserved. An indirect effect is to maximize parental selection probability which participates in the formation of gene structure of the progeny and the next generation in a way that, in the random samples of progeny generation, there is genetic combination which was not in the past generation or their ancestors.

5. Correlation in the silkworm

5.1. Correlation

Every development in animal production improvement depends on the use of breeding methods. By determining the heritability and genetic correlation, the application of appropriate methods for selection and estimating genetic improvement is possible. Previous knowledge and background about heritability of the specific trait is essential for correlation and connection of genes in the females. Before carrying outbreeding experiments, it is necessary to identify how the breeding of a specific trait causes synchronous changes in other traits, and whether in a positive or negative direction (Seidavi & Bizhannia 2008a). In order to investigate and improve all multi-gene traits in every silkworm breeding programme, we must demonstrate correlation information in two ways (Esfandiari et al. 2011): (1) correlation between two or among several quantitative traits in domesticated animals; more of the economic traits have correlation with each other and (2) correlation for one trait in two variant types of domesticated animals. There have been many investigations for estimation of heritability and correlation among economic traits of silkworm to develop and optimize production through selection systems (Seidavi et al. 2004a; Seidavi et al. 2008b). Furthermore, genotypic correlation coefficient is more important than phenotypic correlation. A decrease of phenotypic correlation is connected with various genotype levels of traits as a result of variable effects of the environment (Rahman 1984). Thus, Ghanipoor et al. (2008) in the estimation of economic genetic traits, observed high positive correlation in six commercial lines of silkworm that was similar to the result of Satenahalli et al. (1990). In addition, Kumar et al. (1995) demonstrated that there is a high correlation between cocoon weight and cocoon shell weight, and also between cocoon weight and the per cent of cocoon shell.

5.2. Correlation estimation in silkworm

In the studies of Seidavi et al. (2007) and Vaez Jalali et al. (2011a), genotypicand phenotypic correlations were observed between the whole cocoon production and individual cocoon weight. Moreover, according to the results, for the traits of cocoon shell weight and the cocoon fibre length, selection on cocoon fibre and denier had a positive effect on increase of cocoon production. Positive correlations between the traits of cocoon weight, silken shell weight and pupa weight was reported by Talebi et al. (2001). There was also a negative correlation between cocoon weight and the presence of silken shell of cocoon as reported by Li (1992) and Sohn and Ramires (1999). Various investigations have been accomplished on the amount of heritability of oviparous lines, larvae weight, larvae duration, the growth rate, cocoon weight, pupa weight and cocoon shell weight in silkworm. This shows that correlations among the traits have an important role in silkworm breeding (Sing et al. 1998; Seidavi et al. 2007; Vaez Jalali et al. 2011b). In Seidavi's study (2010a), selection response for the per cent of pupa duration in suitable cocoons in spring was –0.304%and in autumn was –1.486%. Studies have shown that direct selection for one trait is correlated with other quantitative traits. Some have positive correlations and others have been found to have negative correlations.

Pal and Moorthy. (2011) demonstrated that both the length of the larva's body and the length of cocoon had a positive correlation with cocoon shell weight. Anticipated cocoon weight had a significant and positive correlation with cocoon shell weight and also, cocoon shell weight had a significant and positive correlation with cocoon weight and cocoon length. Similarly, the per cent of cocoon shell had a significant correlation with the per cent of shell weight. Grekov (1989) observed that there is a powerful mutual effect between genotype and the environment which causes a positive correlation between cocoon shell weight and cocoon weight (+0.659). Mirhosseini et al. (2010) observed that a high and positive genetic correlation exists between cocoon weight and cocoon shell weight. These two important economic traits in which selection was on cocoon weight caused the increase of cocoon shell weight. Additionally, they observed a decrease of correlation in the Baghdad group and a high correlation in the other groups.

6. Heritability in the silkworm

The principle of quantitative genetics is founded on Mendelic rules, but there are some differences between breeding programmes. Heritability is one of the most valuable parameters in animal breeding. Heritability of one trait in the general sense of quantitative genetics is defined as the proportion of a single trait's variance in a population which is attributed to the average effect of the genes. When genetic and phenotype variance in a population are estimated, heritability can be easily calculated (Talebi et al. 2010). If a silkworm breeding expert selects parent larvae according to their phenotype value, improvement of traits in the next generations can be only anticipated with due attention to the amount of phenotypic difference and the heritability. Therefore, when breeding experts choose the breeding methods and specify the procedure, attention to heritability is essential (Seidavi & Bizhannia 2008a).

This means that each individual gene has one unique composition and genetic effect in the population. So new parents have the most genetic variety in the next generation (Gharahveysi et al. 2010).

Ghanipoor et al. (2006) have investigated the phenotype process for production traits in silkworm under selection condition and demonstrated that phenotypic process and the improvements of economic traits in pure lines are different and influenced by various factors like heritability. Breeding pressure for one trait is a linear function of breeding confidence (Doreswamy & Subramanya 2012). Selection of parental species should exactly conform with commercial needs and necessary production traits because specific parental traits clearly manifest in the offspring (Manohar Reddy et al. 2009). Different breeding systems can balance the traits in line selection for the increase of cocoon efficiency and silk quality (Yamaguchi 2001; Benchamin 2002).

The use of different mathematical and statistical methods increases accessibility to new methods of silkworm selection. Petkov et al. (2007b) demonstrated that actual heritability coefficients in females were significantly greater compared with males which is similar to the results obtained in Petkov's (2007a) studies. As quoted by Ahsan and Ramian (2008), Mirhosseini et al. (2005) have estimated heritability increase of cocoon weight and cocoon fibre weight corresponding to cocoon shell ratio. Singh et al. (1994) have computed a maximum heritability in SSW species 80 of 20% for pupa weight. Seidavi et al. (2003) reported that both cocoon weight and cocoon shell weight are influenced by the increase of energy and predominance variance. Ghanipoor et al. (2006) concluded that heritability of cocoon weight and cocoon shell weight is higher than for the percentage of cocoon shell.

6.1. Estimation of heritability in the silkworm

Variance factors play an essential part in predicting the breeding value, planning breeding programmes, and predicting the genetic process of traits (Sheikhlou et al. 2008). Petkov et al. (2007b), estimated the selection effect in all the lines through statistical methods. The expected breeding value and the actual one for cocoon shell weight trait were parallel. They concluded that genetic selection parameters, actual heritability coefficient, selection difference, mean, standard deviation, and selection intensity have been able to develop a proved scientific programme on the basis of selection on cocoon shell trait and estimation of selection effect. Moreover, Govindan et al. (1991) have reported heritability for cocoon weight trait between 0.03–0.49, cocoon shell weight trait between 0.14–0.60, and the per cent of cocoon shell trait between 0.150–0.9. Mu et al. (1995) found that heritability in cocoon shell weight, the per cent of cocoon shell and the weight of ten thousand larvae is specifically higher than that in other traits. Kumaresan et al. (2000) have estimated heritability in individual cocoon weight, cocoon shell weight and the per cent of cocoon shell respectively as 0.6068, 0.673 and 0.5735. In 2005, Mirhosseini et al. estimated genetic parameters of economic characteristics for domesticated silkworm species and demonstrated that heritability for most economic traits is high.

Kshama et al. (1995) have studied two NB7 and NB4D2 silkworm varieties and reported heritability, genotype and environmental correlation in three competitive traits including fertility, survival, and growth; and also for quantitative traits including cocoon weight, cocoon shell weight, and the length and thickness of cocoon fibre. Heritability in quantitative traits was high, about 48% to 64% and also heritability in competitive traits was lower, from 18% to 25%. Kumar et al. (1995) have studied genetic indicators in the quantitative traits of 46 bivoltine varieties and concluded that heritability of the fibre length, cocoon weight and cocoon shell weight (which is controlled by many gene factors) was high. In addition, the correlation between cocoon weight and cocoon shell weight, and also the correlation between cocoon shell weight and the per cent of cocoon shell were extremely significant. Talebi et al. (2010) report that using the restricted maximum likelihood (REML) method, Mirhosseini et al. (2005) found that the heritability of cocoon weight and cocoon shell weight increased and the per cent of cocoon shell weight in 101,433 lines was 0.73 and heritability of the per cent of cocoon shell in KOMING1 lines was high compared with other lines (0.61).

Nematollahian et al. (2010) studied ten lines and concluded that the highest heritability of cocoon weight and cocoon shell weight was in line 31 and the lowest was in line 153. For the cocoon shell trait, the highest heritability was in line 31 and the lowest was in lines 152, 104 and 32. Additive breeding value for traits including cocoon weight, cocoon shell weight, proportion of shell and fibre length was reported in the studies of Petkov and Nguyenvan (1987) and Maqbool et al. (2005). In the new findings, specific breeding value for cocoon weight, cocoon shell weight, proportion of shell, fibre length and denier was accounted respectively 0.85, 0.53, 0.34, 0.53 and 0.56. A comparison was made between general and specific heritability which demonstrated the lower value of specific heritability. But, estimation of general heritability demonstrated a higher value, and more of these results are influenced by dominance effects existing in the breeding of domesticated animals (Strickberger 1995).

Satenahalli et al. (1990) estimated genetic improvement for cocoon weight and cocoon shell weight at 29.99% and 46.02% respectively. Talebi (2001) evaluated heritability coefficients using the information from offspring of full bloods in three Japanese lines and three Chinese lines. The heritability coefficient for cocoon weight was 0.209 ± 0.123 and for cocoon shell weight it was 0.228 ± 0.129 and for the per cent of cocoon shell it was 0.044 ± 0.042. Seidavi et al. (2010b) in the evaluation of genetic parameters of pure commercial silkworm lines have concluded that heritability for cocoon weight, cocoon shell weight and the per cent of cocoon shell was respectively 0.496, 0.499, and 0.313. Moreover, through several studies heritability of the cocoon weight trait was reported between 0.03 and 0.49 and heritability of cocoon shell weight was reported between 0.14 and 0.60 (Malik 1999; Seidavi 2004b; Seidavi et al. 2008b). By analysing genetic changes in 58 multivoltine silkworm lines in tropical part of India, Kumaresan et al. (2007) have concluded the heritability of cocoon weight, cocoon shell weight and the per cent of cocoon shell traits respectively as 0.4652, 0.565, and 0.3731.

6.2. Voltinism

If the number of males and females selected every year is partly fixed, with the increase of the number of male and female creatures in each generation, generation interval will increase. Therefore, prolongation of generation interval can be an essential method for increasing the size of the effective population and decreasing genetic drift (Gharahveysi et al. 2010).Generation interval refers to the length of time in which the population regenerates (Falconer & Mackay 1996). Although maternal inheritance for the voltinism trait has been proven through several genetic experiments, the circumstance of inheritance is intricate because of the influence of light and temperature in each phase from the growth of the silkworm egg to the butterfly phase. When bivoltine silkworm eggs are maintained at 15°C after the embryo blastokinesis phase, butter flies are produced without winter silkworm egg; in contrast, when silkworm eggs are maintained at 24°C, butterflies are produced through winter silkworm eggs (Seidavi & Bizhannia 2008a). Basically, gene frequency variation is the effect of selection. Usually, three essential factors in quantitative trait improvement are influenced by selection. They include selection difference, inheritance and voltinism (Neshagaran Hemmatabadi et al. 2011b). Genetic variety in productive traits is the basis of artificial selection in animal breeding programmes for genetic improvement of the future population. Genetic variety in traits is influenced by artificial selection in animal breeding programmes (Gharahveysi et al. 2010). Overall, dominance in monovoltine silkworm eggs seems more than in bivoltine; and in bivoltine silkworm eggs seems more than in multivoltine (Seidavi & Bizhannia 2008a).

7. Selection principles in animal science: a brief review

A breeder can change genetic characteristics of a population in two ways. The first way is selecting individuals as parents which means selection, and the second way is controlling the mating of parents which includes inbreeding and outbreeding. All progression in domesticated animal and plant breeding has been achieved through using such methods (Falconer 1990). For many years, animal breeders changed the genetic structure of animals through genetic selection without any information about individual genes. In other words, the main methods for genetic evaluation of animals were chiefly based on phenotype information and pedigree record (Hill & Natt 1990; Goddard 2001). This kind of genetic evaluation is based on infinite genes theory which supposes quantitative traits are influenced by numerous genes having small additive effects and possessing independent segregation. Although the assumptions of this theory are incorrect, most early genetic developments for economic quantitative traits in farm animals was through this kind of genetic model (Falconer & Mackay 1996). Selection is accomplished in two general ways (Falconer 1990): (1) Natural selection, in which nature's power interferes and (2)Artificial selection, which is accomplished by humans. Artificial selection means selection of parents by a breeder that separates mature individuals of the parent species into two selective and cull groups causing changes in gene frequency (Falconer 1990; Soysal 2004).

7.1. Selection principles in silkworm

The success of a breeding programme is measured by the amount of breeding value changes of traits under selection. One of the best available tools for maximizing the response in selection programmes is correct prediction of the breeding value of the next generation's parents (Jurado et al. 1994). Adequate, correct and permanent selection of parents causes quantitative and qualitative improvement of the function of varieties due to breeding methods. These days, selecting parameters such as genetic variance, covariance, correlation, heritability and economic traits can be profitable factors (Falconer 1989). For achieving the goals of selection, we should be able to access the animals having a variety of genetics. For example, adequate layers of silkworm egg possessing proper genetics can extensively change the potential of the sericulture industry (Ramesh 2001; Chandrashekaraiah 2003). In recent decades, numerous studies have been accomplished about the effect of breeding on animals and poultry which have mostly investigated the breeding pressure on the population under study. However, breeding pressure on economic traits of the silkworm population has been ignored. A small population and also individual selection can increase breeding level; therefore, knowledge of the breeding coefficients and their effects on economic traits are essential for estimating the value of breeding the lines (Azizian et al. 2011).

Selection of parent lines for a breeding programme is based on their characteristics (Reddy et al. 1999). Moreover, individual selection in silkworm parents can improve results, but these improvements are influenced by environmental factors – especially the season of the year (Seidavi 2010a; Seidavi 2010b). In addition, intense selection in one line may eliminate specific alleles (Foulley & Ollivier 2006). For specifying the strategy of adequate selection, different factors should be regarded such as the type and the number of traits and their particular economic values; heritability and genetic correlation among the traits, genetic and economic response to selection, the goal of breeding, the type of variety, available facilities, and economic and productive conditions. Then, in the selection of lines, an adequate selection method should be used which accords more with the existing condition. Also, recent investigations have shown that economic traits of the silkworm possess high heritability and phenotype selection on individual traits possesses high efficiency (Shabdini Pashaki 2010).

Among the basic principles of the science of genetic and the main method of silkworm breeding, the use of cross-bred power has had a great part in creating new and productive silkworm strains (Seidavi & Bizhannia 2008a). Ashoka et al. (1990) have investigated genetic characteristics including variance, phenotype and genotype, covariance, general heritability, and genetic improvement in nineteen quantitative traits for four bivoltine silkworm varieties, four single cross-breds and eight double cross-breds. They found that genotypic covariance forms the majority of phenotypic covariance for all traits. In addition, they found high heritability and genetic improvement for larvae weight, the number and the weight of produced cocoon, cocoon weight and its shell weight, and also the length and weight of fibres which indicates additive gene effects. Therefore, they believed that suitable results can be also achieved through one single phenotype selection.

The selection programme refers to the use of a pure line for the improvement of function and economic development; so selection has an essential part in genetic improvement (Seidavi et al. 2009). The studies of Kumaresan et al. (2007) demonstrate that genetic variety has an essential part in multivoltine selection of 58 silkworm strains, with genotype, and estimation of genetic variety of ten economic traits for utilizing as a genetic source in the silkworm breeding programme. Selection of the parent lines is important because so much time and expense is spent experimenting on the population obtained from the line. Basically, production of superior offspring is affected by parent selection, not the number of applied parents. There are three criteria for parent line selection (Seidavi & Bizhannia 2008a): (1) Recognizing variety traits of parents such as larvae period, the per cent of cocoon shell, the length of fibres and the per cent of raw silk, etc., (2) Recognizing superior varieties as parents, behaviour and pattern of oviposition, the number of eggs produced, the survival rate of fertilized eggs, hatching consistency, etc., and (3) Experimental results of cross-breeding capability indicating the general value of a parent. Selection on one indicator of a trait is connected with other quantitative indicators which are economically important. The connection between the problems among heterosis traits, caused by selection or non-selection, is related to the history of selection in a population. Selection of silkworm lines having lengthy fibres demonstrated that the increase of fibre length is connected with the decrease of fibre amount, that is, it causes a decrease in the total amount of fibre (Nagaraju et al. 1995).

7.2. The effects of environment on silkworm selection

The productive function for each variety is defined by genetic effects, environment effects and any interactions between these two. Therefore, to improve the quantity and quality of produced silk, not only should we improve environmental conditions such as nutrition, sanitation, management and rearing condition, but we should also follow fundamental steps for increasing genetic power and breeding (Mavvajpour et al. 2008). However, the selection method is usually dependent on genetic selection and many factors can cover the real genetic capacity of an animal. Therefore, for genetic evaluation it is essential to recognize environmental factors and maintain correctly produced records based on them (Taheri Dezfuli 2004).

Selection of the population varieties are an important basis of a breeding programme. Quantitative economic traits of silkworm are influenced by environmental factors. For a suitable response to selection, it should be recognized that divided populations are affected by additive genetic factors, the ratio of phenotype traits should be omitted and finally, selection should be done based on additive genetic value (Mirhosseini et al. 2008b).

Neshagaran Hematabadi et al. (2011a) studied selection effects on three commercial lines of silkworm (lines 31, 103 and 107)and concluded that in spring, line 31 showed the most response to selection and line 103 showed the least response. In the whole year, line 31 showed the best response and respectively line 103 and 107 showed the least response to selection.

Rangaia et al. (1995) studied about six traits in silkworm genotypes and reported that fertility has the most phenotypic and genotypic variety. Environmental effects have had the most effect on fertility and cocoon shell weight, but, the length of larval period and the per cent of silken shell were comparatively less influenced by environmental factors. In addition, cocoon weight, cocoon shell, the per cent of cocoon shell, and maximum weight of larvae have the highest heritability, and fertility has medium heritability; and the length of larval period has low heritability. It was suggested that breeding programmes be planned based on selection for the traits such as cocoon weight and the maximum weight of larvae because for the sake of expected heritability and genetic importance, they possess maximum response to selection. The effects of parent selection due to cocoon weight on species characteristics and offspring lines should be separately investigated in each country in order to consider management systems according to regional conditions (Seidavi et al. 2008b).

7.3. Accomplished activities based on silkworm selection

Silkworm breeding plays an essential part in the increase of cocoon efficiency and silk quality. Selection is one of the best approaches in silkworm breeding (Seidavi et al. 2008b; Seidavi 2010a). Selection based on quantitative traits, genetic correlation and inheritance has a great importance (Kumar et al. 1995). Seidavi (2010a) stated that individual selection on cocoon weight had no negative effect on important economic and productive traits. Selection is important in animal and plant breeding, and in the formation of genetic varieties from an economic point of view (Hojatpour et al. 2012). Much research has been accomplished in the areas of genetics and breeding and specifically on the effect of individual selection on the improvement of function in silkworm varieties. For example, all studies on the multivoltine lines showed that these varieties possess the highest resistance against changing weather conditions causing a decrease in silk production, in comparison to monovoltine and bivoltine species (Rao et al. 2006). The studies of Hojatpour et al. (2012) show that individual selection of pure parental lines of silkworm have no significant effects on the function of hybrid offspring. Therefore, egg production of commercial lines with individual selection systems can be useful through pure parental lines.

Seidavi (2010a) reported that response to selection for the pupa survival trait in medium cocoon was 0.074 and–2. 627 in spring and autumn respectively. Response to selection for the resistance trait of silkworm in pure lines had a significant increase in 0.05 level in spring in comparison to autumn. An estimation of genetic parameters of pure silkworm lines, Seidavi (2010b) reported that the genetic change was significant and positive in the population on which no selection was done. The genetic process for the cocoon weight trait was 0.016gr for each species, cocoon shell weight was 0.005gr in each species and the per cent of cocoon shell was 0.071% in each species. In this study, phenotypic change of cocoon weight, cocoon shell weight and the per cent of cocoon shell in selected and non-selected groups were reported negative and significant. Phenotype change in non-selected population for traits of cocoon weight, cocoon shell weight and per cent of cocoon shell was respectively–0.061, –0.018gr in each line and –0.314%ineach variety; while in the selected group, it was respectively –0.108, –0.029gr in each line and –0.379% in each variety. Therefore, it shows that the phenotype change in the non-selected group is higher than in the selected group.

Haile Mariam et al. (1990) have investigated the cross-breeding capability of seven silkworm varieties in relation to the traits of hatching, pupation weight, the length of pupation period, butterfly appearance, fertility, cocoon weight and fibre length. Then, they confirmed the general cross-breeding capability of these varieties with each other. In the investigations of Mohapatra (2009) and sing et al. (2009), the effect of environmental factors on the function of silkworm rearing has been studied. In this study, selection caused an increase in efficiency of genotype function in compatible silkworm varieties; this is one of the goals of most breeding programmes.

8. Conclusion

As a conclusion, silkworm breeding could improve economically valuable traits to increase the profits of the sericulture industry. Selection under different environmental conditions, calculation of breeding values, quantitative and qualitative traits, consanguinity and heritability are the most tools for improvement of silkworm productivity.

Acknowledgements

We are grateful to the Rasht Branch, Islamic Azad University for providing us with the necessary support. This work was also supported by Qaemshahr Branch, Islamic Azad University. We thank Dr Jan Horton (Tasmania) for editing the English text.