Characterization of indigenous silkworm (Bombyx mori L.) breeds on the basis of morphological traits

Silkworm (Bombyx mori L.) originated in China in the southern district of Yangtze. Owing to its commercial value; the insect was brought to Korea, Japan and Europe and subsequently to India through the Silk Road. In due course of time, the species adapted to respective local conditions and got differentiated into Chinese, Japanese, European and Korean varieties and thereafter is being domesticated for over 5000 years. It has its use in textile as well as non- textile industries and is a model organism for studying microbiology, physiology and genetics. Silkworm comprises of large number of geographical races and inbred lines; distributed in temperate and tropical countries and showing substantial variation in their qualitative and quantitative traits. Currently, it is the major economic resource for nearly 30 million families in countries such as China, India, Japan, Korea, Vietnam and Thailand (Rathore et al., 2011; 2012). The tropical genotypes (non diapausing) are hardy and withstand adverse eco-climatic conditions but produce small quantum of poor quality silk, whereas genotypes of temperate (dia-pausing) origin produce higher quantity of international quality silk. The conventional breeding approaches, although have increased the silk productivity, but it has not integrated the high yielding traits of temperate genotypes with the low yielding disease resistant tropical genotypes. Moreover, the environmental factors interact specifically with the phenotypic and conventional breeding and, thereby limit the yield improvement of the strains of silkworm. The genetic resources are the basic material for crop improvement and the success of a breeding programme depends on the initial selection of parents and their effective utilization in desirable combinations. Further, it also depends on the ability of breeds to assemble and recombine the genetic variability and extraction of the potential gene combinations from the gene pool based on phenotypic expression; leading to genetic fixation of the traits over generations. Therefore, silkworm genetic resources need to be properly evaluated and screened for identifying the potential of promising parental lines. The earlier works conducted on the characterization of silkworm genetic resources Ram et al., (2003, 2006) and Sajgotra et al., (2016). In light of above background, the present study has been designed to characterize the indigenous silkworm bivoltine breeds on the basis of morphological traits.

Rearing of bivoltine silkworm breeds and qualitative characterization
 
The experimental bivoltine silkworm breeds were characterized for morphology of egg, larva, cocoon, pupa and adult. The characterization was done by collecting the samples at various stages of development. The research work was conducted at the SKUAST-Jammu with 20 different races of indigenous silkworm breeds namely; PO₁, PO₃, SPO, ND₂, ND₃, ND₅, NSP, UDHEY-1, UDHEY-2, UDHEY-3, UDHEY-4, UDHEY-5, UDHEY-6, UDHEY-7, UDHEY-8, CC₁, CSR₁₈, CSR₁₉, NB₄D₂ and SH₆. The rearing of silkworm was conducted at the Division of Sericulture Udheywala, SKUAST-J, with standard rearing techniques suggested by (Dandin et al., 2003). To prevent diseases and to maintain good sanitation, the rearing room and other rearing appliances were disinfected with a 5% formaldehyde solution following Ullah and Narasimahanna (1978) procedure. The eggs were disinfected with 2% formaldehyde solution for five minutes recommended by Jolly (1983) incubated at 25+1°C and 80%+5% humidity and transferred to the rearing house, the newly hatched silkworms transferred to the rearing tray with a feather following the brushing technique of Ullah and Narasimahanna (1978). Cocoons were harvested after 4 to 5 days of mounting. Male and female pupae were segregated and the fresh moths were allowed to mate after emergence. The eggs produced were called disease free layings (dfls) and they were utilized for experiment. The characters studied and observational procedures are given below:
        
Fecundity was calculated by counting the total number of eggs female and mean of three layings in each replicate was observed for analysis and with these hatching percentage was computed. V^th Instar larval duration (D: H) was observed duration from fourth instar moult out upto pre-spinning stage and was recorded in days and hours for each breed; and total larval duration (D;H) from span in days and hours from brushing to pre- spinning stage including the moulting period in each instar. Ten mature larvae were picked randomly from each replicate of each breed from 4^th to 6^th day of fifth instar and weighted on digital balance. Maximum larval weight was recorded in each breed. Pupation % this may be effective rate of rearing (ERR) represents the number of live pupae obtained and calculated from the number of live pupae in cocoons harvested and total number of larvae retained after III moult. Malformed cocoon in the form of double, flimsy and dead cocoons were also computed in percentage. Single cocoon weight (g) was obtained from 25 male and female cocoons which were randomly selected and weighed on digital balance. Likewise single shell weight was also recorded. From these shell ratio % was worked out to determine the average cocoon weight by using the following formula; Weight of 25 male + 25 female cocoons/ 50 × 100.
        
The data obtained were analyzed for means and standard deviation followed by the ranking of parental breeds Mano et al., 1993 using Evaluation Index Method, (E.I.) with the formula: E.I. A – B/ C × 10 + 50 where, A = Value obtained for a particular traits of the breed; B = Mean value of a particular trait of all breed; C = Standard deviation of a particular trait; 10 = Standard unit; 50 = Fixed value and the coefficient of correlation of each character with yield was computed separately and were performed using following formulae:

 

Where, N is equal to the number of pairs of scores and Σxy is called the sum of the cross, followed by Multivariate and clustering analysis. Data from all the considered breeds was estimated using PAST3, which creates cluster analysis of the Euclidian distance metric to form an agglomerative hierarchical tree through UPGMA (Unweighted Pair Group Method using Arithmetic Averages). Tree has been presented as 2D plot Dendrogram. Principal Component Analysis (PCA) and Principal coordinate analysis (PCoA) was determined from coefficients output from Euclidian distance metric using software PAST3. All phenotypic data was used to compute various association coefficients for qualitative traits to form a Euclidian distance metric, which was subjected to make a Double Centre distance metric. Double Centre distance metric was used as input to create Eigen vector, Eigen value and graphical presentation of 2D plots.

Qualitative characterization of bivoltine silkworm breeds
 
The silkworm rearing trial of twenty bivoltine silkworm breeds was carried out in 2017 spring rearing season. The results obtained are presented below.
 
1. Egg shape
 
The shape of the eggs of all silkworm breeds studied, were ellipsoidal type.
 
2. Chorion colour
 
Twelve breeds viz. PO₁, PO₃, SPO, UDHEY-1, UDHEY-2, UDHEY-3, UDHEY-4, UDHEY-5, UDHEY-6, UDHEY-7, UDHEY-8 and SH₆ recorded creamish white chorion colour. Five breeds namely ND₂, ND₃, ND₅, NSP and NB₄D₂ were light yellow in colour. Remaining three breeds of CSR&TI Mysore origin viz. CSR₁₈, CSR₁₉ were having pale yellow colouration while as breed CC₁ had white chorion (Table 1).
 

 
3. Colour of newly hatched larva
 
Out of twenty breeds, ten breeds viz. UDHEY-1, UDHEY-2, UDHEY-4, UDHEY-5, UDHEY-6, UDHEY-8, CSR₁₈, CSR₁₉, NB₄D₂ and SH₆ recorded dark brown colour. Four breeds PO₁, PO₃, SPO and CC₁ were brown in colour while as remaining six breeds viz. ND2, ND3, ND5, NSP, UDHEY-3 and UDHEY-7 were having black neonatal larval colouration (Table 2).
 

 
4. Larval marking
 
The larval marking in nine breeds; PO₁, PO₃, SPO, UDHEY-2, UDHEY-4, UDHEY-6, UDHEY-8, CC₁ and NB₄D₂ were plain and breeds viz. ND₂, ND₃, ND₅, NSP, UDHEY-1, UDHEY-3, UDHEY-5, UDHEY-7 and SH₆ had crescent marked larvae. Two breeds, CSR₁₈ and CSR₁₉ were sex limited having marking on female larvae only while as male larvae were plain with no markings (Table 3).
 

 
5. Cocoon Colour
 
Among twenty breeds, eight breeds viz. PO₁, PO₃, ND₅, UDHEY-3, UDHEY-4, UDHEY-5, UDHEY-6 and UDHEY-8 recorded creamish white colour while as eleven breeds; SPO, ND₂, ND₃, UDHEY-1, UDHEY-2, UDHEY-7, CC₁, CSR₁₈, CSR₁₉, NB₄D₂ and SH₆ were white in colour. Breeds NSP recorded dull white coloured cocoons (Table 4).
 

 
6. Cocoon shape
 
Among twenty breeds studied, nine breeds viz. PO₁, PO₃, SPO, UDHEY-2, UDHEY-4, UDHEY-8, CC₁, CSR₁₈ and SH₆ had oval cocoon shape. Seven breeds viz. ND₂, ND₃, UDHEY-1, UDHEY-3, UDHEY-5, CSR₁₉ and NB₄D₂ had elongated faint constriction and breed UDHEY-6 recorded oval faint constriction. Three breeds, viz. ND₅, NSP and UDHEY-7 were elongated constricted (Table 5).
 

 
7. Pupa colour
 
Ten plain larva viz. breeds PO₁, PO₃, SPO, UDHEY-2, UDHEY-4, UDHEY-6, UDHEY-8, CC1, CSR_18, CSR₁₉ (male) and NB₄D₂ recorded yellow colour while as ten marked larvae ND₂, ND₃, ND₅, NSP, UDHEY-1, UDHEY-3, UDHEY-5, UDHEY-7, CSR₁₈ CSR₁₉ (female) and SH6 were of yellowish brown colour. The phenotypic pupal colouration for plain and marked larvae was in 1:1 ratio (Table 6).
 

 
8. Pupa shape
 
No variation in pupa shape was observed and all the pupae of all twenty breeds were of elliptical shape.
 
9. Colour of male and female moths
 
No variation in male moth colour was observed. All the twenty breeds studied had creamish white coloured male moths. Limited variation in female moth colour was observed. Female moth colour in breeds ND₂, ND₃, ND₅, NSP, UDHEY-1, UDHEY-2, UDHEY-3, UDHEY-4,, UDHEY-5, UDHEY-6, UDHEY-7, UDHEY-8, NB₄D₂ and SH₆ were dirty white while as breeds PO₁, PO₃, SPO and CC₁ had dull white female moths. Sex limited breeds, CSR₁₈ and CSR₁₉ had dirty colour female moths (Table 7).
 

 
Quantitative Characterization of bivoltine silkworm breeds
 
The performance of twenty breeds is depicted for their parameters at egg, larval and cocoon stages in fecundity was observed to be maximum with the breed CSR₁₉ (639) followed by UDHEY-1, UDHEY-3 (574) and the least in NB₄D₂ (476). The hatching percentage was maximum in SPO (96.28) followed by UDHEY-2 (94.88) and minimum in breed ND₂ (80.90). Among breeds, fifth age larval duration was longer in PO₁, ND₂, ND₃, ND₅, NSP, CC₁, CSR₁₈, CSR₁₉, NB4D₂ and SH₆ (7:0) and shorter in PO₃, SPO, UDHEY -1, UDHEY-2, UDHEY-3, UDHEY-4, UDHEY-5, UDHEY-6, UDHEY-7 and UDHEY-8 (6:0). The longest and above average total larval life was recorded in breed NSP (29.00) followed by CSR₁₈, CSR₁₉ and NB₄D₂ (28.00). Shortest total larval duration was observed in breeds UDHEY-1, UDHEY-2, UDHEY-3, UDHEY-4, UDHEY-6 and CC₁ (26.00). Maximum single mature larval weight was randomly recorded in breeds; PO₁, PO₃, ND₂, ND₃ and UDHEY-3, (3.7) followed by SPO, ND₅, NSP, UDHEY-5, CSR₁₈ and CSR₁₉ (3.6) whereas, it was minimum in UDHEY-1, UDHEY-6, CC₁, NB₄D₂ and SH₆ (3.4). The percentage of malformed cocoons were maximum in breed PO₁ (3.98) followed by SH6 (3.79) and least in breed UDHEY-6 (2.20). The malformed cocoon comprises of flimsy, thin end and deformed cocoons. Highest pupation percentage was recorded in breed ND5 (92.00) followed by ND₃ and NSP (90.00). It was lowest in breeds UDHEY-6 and NB₄D₂ (82.00). Cocoon weight was maximum in breed UDHEY-3 (1.800) followed by ND₃ (1.726) and PO₃ (1.725). Breed UDHEY-6 recorded minimum weight of 1.425g. Three breeds PO₃, ND₃ and ND₅ recorded higher single shell weight of 0.360 followed by breeds PO₁, SPO, ND₂ and UDHEY-3 (0.350). Lowest value of 0.290 shell was recorded in breed NB₄D₂. Highest and above average silk % was observed in breed UDHEY-6 and CC₁ (21.75) followed by UDHEY-7 (21.44) ND₅ (21.35) UDHEY-4 (21.02) are presented in Table 8.
 

        
For assessing superior breeds on cumulative basis Evaluation Index method was adopted and results are presented in Table 9 and 10. The evaluation index values for fecundity, hatching percentage, larval weight for 10 mature larvae, malformed cocoon, pupation percentage, single shell weight, single cocoon weight and silk per cent. Maximum Evaluation Index value of 77.52 was observed in breed CSR₁₉ followed by UDHEY-I and UDHEY-3 (60.17) and the least value was in NB₄D₂ (34.00) based on fecundity. Based on hatching E.I. value was maximum in SPO (71.10) followed by UDHEY-2 (66.42) and CSR₁₉ (63.84). Based on larva weight of 10 mature larvae on 5^th (D), E.I. value was maximum (63.63) in PO₁, PO₃, ND₂, ND₃ and UDHEY-3 and based on the 5^th Instar larval duration (D) highest E.I. value of 59.80 was recorded in PO₁, ND₂, ND₃, ND₅, NSP, CC₁, CSR₁₈, CSR₁₉, NB₄D₂ and SH₆ indicating longer fifth age. Based on total larval duration (D), maximum E.I. value of 75.00 was obtained in breed NSP followed by CSR₁₈, CSR₁₉, and NB₄D₂ with E I value of 62.80. Based on malformed cocoon percentage, maximum E.I. value of 67.14 was obtained in breed PO₁ followed by SH₆, (63.26). Based on pupation rate percentage, maximum E.I. value of 70.40 was recorded in ND₅ followed by ND₃ and NSP with E.I. value 63.05. Based on single cocoon weight (g), maximum E.I. value of 70.22 was observed in UDHEY-3 followed by ND₃ with E.I. value of (62.00). Based on single shell weight (g), maximum E.I. value of 64.00 was recorded in PO₃, ND₃ and ND₅ followed by PO₁, SPO, ND₂ and UDHEY-3 (59.00). Based on silk percent, maximum E.I. value of 63.59 was recorded in UDHEY-6 and CC₁ scored followed by UDHEY-7 at 60.11. The important commercial traits of silkworm on which silk industry sustains are fecundity, hatching, larval weight 5^th day of 5^th instar, 5^th instar duration, total larval period, percentage of malformed cocoons, pupation rate, single cocoon weight, single shell weight and silk percent, the cumulative E.I. value for 5^th instar and total larval life was maximum in NSP with E.I value of 67.40 followed by CSR₁₈, CSR₁₉ and NB₄D₂ at 61.30. Breed PO₁, ND₂, ND₃, ND₅ and SH₆ recorded cumulative E.I values of 55.20. The cumulative E.I value for eight commercial parameters recorded maximum E.I value of 58.48 by ND₃ followed by ND₅ (58.37) and CSR₁₉ (56.61) are presented in Table 10.
 

 

 
Linear Pearson Correlation analysis between parameters
 
Phenotypic data of ten parameters was analyzed by Pearson correlation coefficients (Table 11). Highly significant and positive correlations were observed between hatching percentage and 5^th day larval weight (0.545). Single cocoon and shell weight with hatching percentage (0.551) and (0.492) respectively. Pupation rate with 5^th day larval weight (0.524), single cocoon weight with 5^th day larval weight (0.864). Single cocoon weight with 5^th day larval weight (0.816). Total larval duration with V^th instar duration (0.559). Malformed cocoon percentages with duration of V^th instar (days) (0.730). Single cocoon weight with pupation rate percentage (0.536), single shell weight with pupation rate percentage (0.624). Single shell weight with Single cocoon weight (0.775). Silk percentage with single shell weight (0.470).
 

 
Fecundity
 
While studying the correlations among ten economic parameters fecundity parameter showed positive correlation (P<0.01) while as with silk per cent the correlation was negative and non-significant.
 
Hatching percentage
 
Fifth day larva weight (0.545) had significant positive correlation with single cocoon weight (0.551) and shell weight (0.492).
 
Duration of V^th Instar
 
This character correlated positively with total larval duration (0.559) and malformed cocoon percentage (0.730).
 
Total larval duration
 
The total larval duration parameter was non-significant.
 
Matured larval weight
 
Mature larval weight had significant and positive correlation with pupation rate (0.524), single cocoon (0.865) and single shell weight (0.816).
 
Malformed cocoon percentage
 
This character was also non- significant and had no correlation with any parameters.
 
Pupation rate percentage
 
Pupation rate depicted significant correlation with single cocoon (0.536) and single shell weight (0.624).
 
Single cocoon weight
 
Single cocoon weight was positively correlated with single shell weight (0.775).
 
Single shell weight
 
Single shell weight had significant correlation with silk percentage (0.470; P<0.01).
 
Multivariate analysis on phenotypic traits
a. Euclidean distance metric analysis
 
The Euclidean distance metric among breeds considering all the ten parameters are presented in Table 12. The results show that higher the distance, greater the phenotypic divergence. In analyzed data, maximum distance was observed between NB₄D₂ and CSR₁₉ (163.25) followed by CSR₁₉ and UDHEY-5 (145.13), CSR₁₉ and PO₃ (142.15), CSR₁₉ and CC₁ (136.28), CSR₁₉ and UDHEY-8 (132.28), CSR₁₉ and UDHEY-7 (128.19), CSR₁₉ and UDHEY-2 (128.14), CSR₁₉ and ND₂ (124.12), CSR₁₉ and UDHEY-6 (117.63), CSR₁₉ and SPO (108.16), CSR₁₉ and UDHEY-4 (104.51) whereas minimum distance was observed between UDHEY-1 and UDHEY-3 (4.20) followed by UDHEY-7 and  ND₂ (4.64), UDHEY-5 and PO₃ (5.43) and UDHEY-8 and  UDHEY-7 (5.48).
 

 
b. Dendrogram analysis on phenotypic traits
 
The dendrogram obtained on the basis of phenotypic data elucidates the genetic relationships among breeds, considering all ten traits (Fig 1). All breeds were grouped into two clusters, Cluster A and Cluster B. Cluster A was sub divided into two sub-clusters namely cluster A-I and cluster A-II. Cluster A-I containing eight breeds PO₁, SH₆, NSP, ND₅ UDHEY-3, UDHEY-1, CSR₁₈ and ND₃. Cluster A-II containing eleven breeds UDHEY-6, UDHEY-4, SPO, UDHEY-5, PO₃, CC₁, UDHEY-8 UDHEY-2, UDHEY-7, ND₂ and NB₄D₂. Cluster B containing only one breed CSR₁₉ (Table 13).
 

 

 
Principal component analysis (PCA) and Principal coordinate analysis (PCoA)
 
Principal Component Analysis (PCA) and Principal Coordinate Analysis are sort of multivariate analysis utilized in present study, where components and coordinates representing different axis of differentiation and the amount of variation accounted for by each of such axis, respectively were derived for twenty breeds of silkworm. These PCA and PCoA were used to obtain two dimensional pictorial representations of all the silkworm breeds. Genetic divergence was measured between the breeds in terms of spatial distribution of targeted genomic region presented in Fig 2 and 3. Genetic divergence was found maximum between NB₄D₂ and CSR₁₉ whereas minimum distance was found between UDHEY-7 and ND₂. Breeds CSR₁₈ and CSR₁₉, CSR₁₉ and SPO, UDHEY-6 and SPO, UDHEY-6 and ND5, PO₃ and UDHEY-2, UDHEY-5 and UDHEY-7, CC₁ and ND₂, UDHEY-8 and SPO, UDHEY-4 and CSR₁₈, SH₆ and UDHEY-1, UDHEY-6 and UDHEY-1, UDHEY-4 and UDHEY-1 exhibited high genetic diversity, whereas breeds, PO₃ and UDHEY-5, UDHEY-3 and UDHEY-7, UDHEY-7 and ND₂, UDHEY-3 and CSR₁₈, ND₅ and NSP, ND₃ and ND₅, UDHEY-8 and CC₁, UDHEY-6 and UDHEY-4 expressed very low genetic divergence with Eigen values depicted in Table 14.
 

 

 

        
PCoA was calculated by computing the distance of interval data from distribution metric. PCoA results depicted CSR₁₉ and SPO having greater divergence, whereas it was minimum between UDHEY-7 and ND₂ followed by UDHEY-7 and UDHEY-2. Breeds NB₄D₂ and SPO; CSR₁₉ and SPO; SPO and NB₄D₂; ND₅ and NB₄D₂; NSP and UDHEY-2; UDHEY-2 and UDHEY-6; UDHEY-6 and ND₅; UDHEY-1 and SPO; UDHEY-3 and SPO; CC₁ and CSR₁₉; SH₆ and NB₄D₂; UDHEY-6 and SPO and UDHEY-4 and SPO, exhibited high genetic diversity. Breeds UDHEY-1 and CSR₁₈; ND₃ and UDHEY-3; CSR₁₈ and PO₁; UDHEY-8 and CC₁; UDHEY-5 and PO₃; ND₅ and NSP and UDHEY-4 and UDHEY-6, showed very low genetic divergence, with Eigen values shown in Table 14.
        
Morphological characters are predominant descriptors for assessment of breeds for distinctiveness, uniformity and stability; but are always influenced by prevailing environment. In present study, the morphological characters were recorded wide phenotypic variation in chorion colour, colour of newly hatched larvae, larval marking, cocoon colour, cocoon shape, pupa colour and moth colour except egg and pupa shape. The observations recorded are in line with earlier workers, Basavaraja et al., (2005); Ram et al., (2006) and Sajgotra et al., (2016) and two patterns of larval marking, plain and crescent were observed. Plain larvae did not have any markings, whereas crescent shaped marking had different characteristic spots. In literature, different type of larval markings have been reported, these include zebra, speckled, quail, multistar and others (Tazima 1978). Larval marking is useful variable, as it helps to prevent the mixing of larvae of different strains. Cocoon shape is an extremely relevant variable in commercialization, since automated wiring admits only elliptical cocoons into the machines (Basavaraja et al., 2005). However, the breeds evaluated in the present study were oval, elongated with faint constriction and elongated constricted. All strains produced white cocoons that varied in brightness indicative of qualitative distinction. The egg laying potential of B. mori. has been noticed to be a heritable character expressed within the genotypic limitations and the results revealed maximum fecundity in breed CSR19 followed by UDHEY-3. The superior fecundity indicates their genetic constitution. Hatching percentage is an important component reflecting variability of the eggs and higher hatching percentage in breed SPO indicates the genetic and physiological state of the female moth. The results fall in line with results of earlier workers Ram et al., (2003 and 2006). Larval duration is an important attribute of economic value and shorter larval duration minimizes the quantum of total food consumption and labour requirement. In the present study, the total larval duration was longer in higher metric trait breeds in comparison to UDHEY breeds. The results are in accordance with Ram et al., (2003, 2006). Daniel et al., (2015) reported that rate of development depends on both genetic and environmental factors. Larval weight is one of the important parameters which determine not only the health of the larvae but also the quality of cocoons spun. Non-significant variation in larval weight of studied breeds was recorded. The present findings confirm the observation made by Masarat et al., (2014). Cocoon weight, shell ratio and filament length are highly heritable traits determining the quality, quantity and efficiency of the reeling. The observations made for five important parameters of cocoon stage revealed significant variations among breeds. Higher cocoon weight was recorded in breeds UDHEY-3, ND3, and PO3 indicating clear difference in maximum nutrient utilization by the breeds in 5^th instar. The present findings are in agreement with Masarat et al., (2014). Present findings are also in conformity with the report of Singh et al., (2010) who concluded that environmental factors influence the physiology of the insect and also have deleterious effect on the economic traits. Cocoon shell weight is an important character in determining the silk content. Present results are in conformity with the findings of Basavaraja et al., (1998) and Sajgotra et al., (2017) who have reported that cocoon shell weight shows variability under varying environmental conditions. The variations in the present finding in the shell weight and silk percent may also be due to racial character.
        
The degree of association or relationship between two variables is measured by correlation coefficient. The correlation coefficient can be positive or negative. Positive correlation indicates that the two variables are varying in the same direction. In silkworm, there are very intimate correlations between some of the characters and an excellent character sometimes brings down the merit of other correlated character. Correlation coefficient provides the estimates of degree of association between two or more traits. In B. mori. although yield is the trait of prime importance, association of other economic traits of value also helps in improving the efficiency of selection.
        
In present study, highly significant and positive correlations were recorded between single cocoon weight and 5^th day larva weight (0.864) followed by single shell weight and 5^th day larval weight (0.816), single shell weight and single cocoon weight (0.775), malformed cocoon and 5^th instar duration (0.730), single shell weight and pupation rate (0.624), total larval duration and 5^th instar duration (0.559), single cocoon weight and hatching percentage (0.551), larval weight 5^th day and hatching percentage (0.545), single cocoon weight and pupation rate (0.536), pupation rate and larval weight 5^th day (0.524), single shell weight and hatching percentage (0492) and silk percent and single shell weight (0.470). These findings are in accordance with the results of Chatterjee et al., (1993 a), Kamili (1996), Kumaresan et al., (2000); Sonwalkar (2001); Chanda et al., (2013) and Chandrakanth et al., (2016).
        
The Euclidean distance metric between breeds depicted maximum distance between NB₄D₂ and CSR₁₉ (163.25) followed by CSR₁₉ and UDHEY-5 (145.13), CSR₁₉ and PO₃ (142.15), CSR₁₉ and CC₁ (136.28), CSR₁₉ and UDHEY-8 (132.28), CSR₁₉ and UDHEY-7 (128.19), CSR₁₉ and UDHEY-2 (128.14), CSR₁₉ and ND₂ (124.12), CSR₁₉ and UDHEY-6 (117.63), CSR₁₉ and SPO (108.16), CSR₁₉ and UDHEY-4 (104.51) while minimum distance was observed between UDHEY-1 and UDHEY-3 (4.20). The Euclidean distance metric revealed relatively high genetic variation among all twenty silkworm breeds. The considerable phenotypic variations illustrated the genetic diversity among studied breeds and the results are in accordance with reports of Srivastava (2008); Bojan et al., (2007); Nezhad et al., (2010) and Moorthy et al., (2015). UPGMA Dendogram, PCA and PCoA grouped twenty breeds into two main clusters Cluster A and cluster B. Cluster A was sub divided into two sub-clusters A-I and A-II. Cluster AI comprised of eight breeds and cluster A-II eleven breeds. Cluster B comprised of only one breed (CSR₁₉). Similar pattern of clustering observed by Chatterjee and Pradeep (2003); Awasthi et al., (2008); Furdui et al., (2010); Pal and Moorthy (2011) and Gurjar et al., (2018) and the clustering of breeds fall in line with earlier workers.

On the basis of Euclidean distance metric, the performance of the breeds; CSR19, NB₄D₂, PO₃, UDHEY-8, SPO, UDHEY-4, ND2, CC₁, UDHEY-2, UDHEY-5, UDHEY-7, UDHEY-6 have been identified as promising breeds and are recommended for further breeding to boost bivoltine silk production especially in sericulture zones of northern India.

The first author sincerely acknowledges Professor and Head of Department Division of Sericulture SKUAST-Jammu, for providing facilities and support for carrying out research.