Diversity of local goats in Kerala, India, based on morpho-biometric traits

J Jimcy, K C Raghavan and K S Sujatha

College of Veterinary and Animal Sciences, Kerala Agricultural University,
Thrissur, Kerala,India - 680 651
jimcyjoseph@yahoo.com

Abstract

Local goat populations have a very valuable genetic potential for sustainable agriculture. The aim of this study was to develop a profile of the local goats in Kerala based on its morpho-biometric characteristics. Data were collected from four different geographical locations in Kerala, viz. Kozhikode , Thrissur , Kottayam and Trivandrum.

Based on physical traits, the populations were not very distinct and variations were seen with respect to coat color pattern, presence or absence of horns, tassels or beard and hair pattern.  The biometrical traits observed in the present study were body measurements, peak yield and prolificacy. Discriminant analysis based on morphometric measurements revealed that the most discriminative variables were head width and body length, followed by shin circumference and rump length.  Discriminant analysis based on body weight, peak yield and prolificacy revealed that only body weight and peak yield variables have significant discriminative capacity. The distance value (D2) calculated between populations revealed that the goat population of Thrissur district was more distant from all other populations.  Cluster analysis using the above three variables indicated that the goat populations of Trivandrum, Kottayam and Kozhikode formed one cluster.

Key Words: Body measurements, discriminant analysis, peak yield, prolificacy

Introduction

Goats, like other livestock species are recognized as important components of world biodiversity. India is a rich repository of goat germplasm being home to 20 recognized breeds of goat with a total population of 120.10 million (Livestock Census 2003).  Indian goat breeds exhibit enormous variations in growth, fecundity, production of meat, milk and fiber, disease resistance and heat tolerance.  Goat production has witnessed excellent growth over the years despite a negative campaign against it for its perceived adverse impact on vegetation, forest and gazing lands.  However, goat population in Kerala has been showing a declining trend after 1996, mainly due to indiscriminate slaughter, shrinking of grazing lands and urbanization. Of late goat production programmes are getting a momentum, possibly because of the efforts of local self governments. The present goat population in Kerala is 12.13 lakhs (Livestock Census, 2003), which constitutes 34.8 percent of the livestock population in the state and 1.01 per cent of the total goat population in India.

 Goat production in Kerala(Latitude 10°00’N, Longitude 76°25’E)is mainly dependent on its native breed: Malabari or Tellichery, which is supposed to have originated centuries ago by mixing of native feral goats with Arab, Surti and Mesopotamian goats along with the native goats of Western Coast (Kaura 1952).  The breed is highly variable in both physical and biometric characters, owing to its origin. They are medium sized animals.  Coat colour varies widely from completely white to completely black. Thirty one percentage of the goats have long hair.  Males and a small percentage of females (13%) are bearded.  Both sexes have small slightly twisted horns directed outward and upward.  Ears are medium sized, directed outward and downward. (Acharya 1982). Only 12 per cent of animals have tassels (AICRP Research Report, 2005) It is well known for high prolificacy, milk yield and adaptability to the hot humid conditions prevalent in the state. However, not much information is available on the genetic and phenotypic variability of local goat populations in other parts of the state, since most of the studies have concentrated on Malabari goats in Northern Kerala. Although these local populations appeared to provide optimal material for improvement, because of their rusticity and their phenotypic and assumed genetic variability, no concerted efforts have been made to study and describe these local populations despite their roles in rural people’s livelihood. Hence, a study was conducted to characterize local goat populations in Kerala, based on physical and biometrical traits. Since the genetic resources required for the future are difficult to predict, characterization of these local goat populations can make a major contribution to an effective long-term management of the species.

Materials and methods

A random sample of 400 adult female goats, 100 each from four different geographical locations in Kerala, viz. Kozhikode (Latitude11°15’N, Longitude 75°49’E), Thrissur (Latitude10°30’N, Longitude 76°15’E), Kottayam (Latitude 9°36’N, Longitude 76°34’E) and Trivandrum (Latitude 8°29’N, Longitude 76°59’E) formed the material for this study. Data were collected by surveying methods using a structured questionnaire, onsite observation and physical measurements of goats. The morphological characters observed were coat color, presence of horns/ tassels/ beard, ear size, and hair pattern. The peak yield and prolificacy of animals during the year 2006-2007 were recorded. Milk production was recorded at weekly intervals starting one week after kidding until the 12 week of lactation. Milk yields were measured using a 500ml measuring cylinder. The peak value for milk yield was noted for each doe.  The following body measurements were taken using a flexible measuring tape while the animal standing on a leveled surface.  The reference points are given below:

1. Head length: Distance from poll to nostril.

2. Head width: Distance between the outer canthus of right and left eye.

3. Height at withers: Distance from point of withers to toe region.

4. Chest depth: Distance from point of withers to chest floor/point of elbow.

5. Chest girth: Chest circumference / heart girth.

6. Shoulder point width: Distance between the right and left shoulder point.

7. Rump length: Distance from point of ischium to pin bone.

8. Rump width: Distance between the point of ischium.

9. Shin circumference: Canon bone perimeter.

10. Body length: Distance from point of shoulder to pin bone.

Prediction of body weight from body measurements

Data on chest girths, height at withers, body lengths and body weights of 100 adult female goats were collected from the records of All India Coordinated Research on Malabari goats.  A multiple regression analysis was carried out to describe the relationship between the independent variables consisting of body length, chest girth and height at withers of adult females on the one hand and the live weight as the response variable on the other hand.  In the present study, chest girth exhibited high and significant (p<0.001) phenotypic correlation to body weight (0.817) compared to body length (0.612) and height at withers (0.410) in adult female goats.  Prediction equations were obtained using the backward elimination regression procedure of SPSS (Statistical Package for Social Sciences). Linear, logarithmic, inverse, quadratic, cubic, power, growth and exponential functions of chest girth were fitted against body weight by the least squares method.  This allowed the possible inclusion of curvilinear functions in the prediction equations.  Highest coefficient of determination (R²= 0.705) was obtained for prediction equation based on power function of chest girth.

The prediction equation developed is presented below:

Y= A x^b

Where,

Y= Body weight in kilograms

A = Intercept (0.0416)

x= Independent chest girth measured in centimeters

b =Regression coefficient for the power function of chest girth (1.5301)

This equation was used for predicting the body weight of animals from different geographical areas.

Statistical analyses

 For analysis of data SPSS statistical package was used.  For the mean comparison of each biometrical characteristic for different populations one way analysis of variance was used and the pairwise comparison of means was done by using Duncan Multiple Range Test. Canonical discriminant analysis was performed for identifying the set of biometrical characters that best discriminate the populations.  Hotelling’s T² value was found using the program MSTATC and the Mahalanobis distance (D²) between the studied four goat populations was estimated using the formula:
 

D²    = (N₁+ N_2)*T²/(N₁*N₂)    

Cluster analysis of the four goat populations based on biometrical characteristics was performed as per agglomerative hierarchical clustering (Chatfield and Collins 1980).

Results

More than 70 per cent of animals studied in all these goat populations had white coat color or a combination of white with either black or brown, and were predominantly horned (Table 1).  Percentage of animals with tassels was high in Kozhikode goat population.  Beard was absent in majority of animals studied.  Ears were of medium type.  Short haired animals were predominantly seen in all goat population except in Thrissur, where a higher percentage (48%) of medium and long haired varieties was observed.

 Morphometric measurements

For comparison of each morphometric variable among different populations, one way analysis of variance was done (Table 2).

 Discriminant analysis based on morphometric traits 

Multivariate analysis of variance performed taking ten morphometric variables (head length, head width, height at withers, chest depth, chest girth, shoulder point width, rump length, rump width, shin circumference and body length), revealed that all the four goat populations were significantly different.  Hence canonical discriminant analysis was performed using SPSS statistical package.  The most discriminant variables selected by stepwise procedure were, head width, height at withers, chest depth, rump length, rump width, shin circumference and body length.  Analysis showed that the two first canonical variables represented a cumulative total of 89.9 % of total variation.  Variables head width and body length were the most discriminative in canonical correspondence with the ordinate axis (CAN1) and the shin circumference and rump length with coordinate axis (CAN2). 

The Mahalanobis distance (D²) estimated between the goat populations according to the morphometric variables studied is presented in Table 3.  Greatest distance (D²) value was obtained between Thrissur and Trivandrum populations (9.13866), while the Kottayam and Kozhikode populations had the least distance value (3.04962).  On doing cluster analysis, it was found that Trivandrum, Kottayam and Kozhikode populations formed a group and Thrissur formed a separate group.

Body weight

Body weights of adult female goats were predicted using the regression equation based on the power function of chest girth.  At 18 to 24 months, the mean body weights, in Trivandrum, Kottayam, Thrissur and Kozhikode goat populations were 29.2 ± 0.66, 28.2 ± 0.45, 22.9 ± 0.41 and 26.9 ± 0.56 kg, respectively.  At 24 to 36 months, the values for body weight were 29.9 ± 0.52, 29.6 ± 0.62, 25.7 ± 0.82 and 28.0 ± 0.87 kg, respectively.  Weight recorded at 36 to 48 months in the respective populations were 32.3 ± 0.63, 30.3 ± 0.79, 26.6 ± 0.56 and 30.3 ± 0.75kg, respectively.  The body weights above 48 months, were 32.8 ± 0.64, 30.9 ± 0.69, 29.4 ± 0.72 and 30.7 ± 0.86 kg, respectively. 

Peak yield

For comparison of mean values of peak yield in the four goat populations, one way analysis of variance was done, and it was found that there was significant variation between peak yields among different populations (Table.4). 

Prolificacy

Highest percentage of multiple births was recorded in Trivandrum (68) followed by Kozhikode (61), Kottayam (58) and Thrissur (55).  Single births in Trivandrum, Kottayam, Thrissur and Kozhikode goat populations were found to be 32, 42, 45 and 39 per cent, respectively.  The percentage occurrence of twins was higher in Trivandrum (58) when compared to Kottayam (46), Thrissur (48) and Kozhikode (52).  The percentage of triplets born in Trivandrum, Kottayam, Thrissur and Kozhikode was found to be 10, 11, 7 and 8, respectively.  Only one percentage birth of quadruplets was recorded in Kottayam and Kozhikode populations, while none was observed in Trivandrum and Thrissur populations. 

Discriminant analysis based on body weight, peak yield and prolificacy:

Multivariate analysis of variance performed taking body weight, peak yield and prolificacy revealed that all the four goat populations were significantly different.  Hence discriminant function analysis was carried out with these three variables using the statistical package SPSS.  It was found that variables body weight and peak yield have significant discriminative capacity.  Body weight accounted a maximum 84.6 percent of total variation with a canonical correlation value of 0.508, while peak yield contributed 15.3 per cent of total variation with a canonical correlation value of 0.243. 

The Mahalanobis distance estimated between the four goat populations according to the variables studied is presented in Table 5.    Greatest distance value (D2) was obtained between Thrissur and Kottayam goat populations (2.30842), while the least distance recorded between Kottayam and Kozhikode populations (0.19238).   On doing cluster analysis it was found that Trivandrum, Kottayam and Kozhikode formed one cluster and Thrissur a different cluster.

Discussion

Majority of goats studied in all the regions have a color pattern of white and a combination of white with either black or brown.  It was also observed that the color pattern reported for goats of Trivandrum, Kottyam and Trichur is similar to that reported for Kozhikode goats (Malabari breed).  Similar colour pattern for Malabari goats were reported by Raghavan et al (2004).  This may be because of two main reasons. Firstly the local goat population, which existed in the state, had a range of colors from white to black.  Secondly, the predominantly white colored Malabari goats might have freely mixed with these goats resulting in the present colour pattern.  This together with the fact that higher percentage of white coat color offers a better resistance to heat stress in environments characterized by high solar radiation (Hensen 1990) might have offered the white dominated colour pattern a selective advantage.The study revealed high frequencies of horned condition (79.8%) among all the goat populations studied.  Relatively small percentages of goats possessed tassels (9.25%) and beard (9.75%).  The results were in conformity with other reports in Malabari goats. Malabari goats are generally horned and only a small percentage of females are bearded (Acharya 1982).  Ears were of medium type.  Short haired animals were predominantly seen in all goat population except in Thrissur, where a higher percentage (48 per cent) of medium and long haired varieties was observed. Based on physical traits the populations were not very distinct and variability was seen with respect to coat colour, horn pattern, presence or absence of tassels and beard and hair pattern.  This indicates that there was free mixing of all these populations and no population can be classified as a distinct group based on physical characters.

Biometrical characters

The goats of Trivandrum district had significantly higher body weight and larger body dimensions than all other populations except for shin circumference and rump width.  Animals of Thrissur district had the lowest body measurements and the populations of Kottayam and Kozhikode came in between. Though body weight was predicted using the regression equation based on power function of chest girth alone, the animals having higher predicted body weights had higher values for other measurements also.  Similar trend for body measurements were also reported by Herrera et al (1996). Animals of Trivandrum and Kottayam had higher body weight than Malabari goats from Kozhikode. This difference might have arisen due to the effects of environmental factors and selection pressure. . Highest mean value for peak yield was recorded in goat populations of Kottayam (833 ± 35.1ml), followed by Kozhikode (755 ± 38.5ml) and Trivandrum (700 ± 40.3ml), while the lowest value was recorded in Thrissur (444 ± 25.4).  Kottayam animals had higher average than Malabari goat of Kozhikode.  A peak yield of 654 ± 46.3 was reported for Malabari goats (AICRP Research Report 2005).

The percentage of multiple births was higher in Trivandrum (68percent) compared to other populations, indicating higher prolificacy for animals of this area.  The percentage of twin births was higher compared to singles, triplets and quadruplets in all the populations studied. Similar observations were made by Shanmughasundaram (1957), Raghavan et al (2004), Bindu (2006) and Seena (2006) in Malabari goats.  High prolificacy is considered to be a character of local goat populations.

Discriminant analysis

The most discriminant variables selected by stepwise procedure were, head width, height at withers, chest depth, rump length, rump width, shin circumference and body length. Variables that contribute to differentiate populations were similar to those found by Herrera et al (1996) in Andalusian goat breeds, Crepaldi et al (2001) working on goat populations from Lombardy Alpines and Lanari et al (2003) in local Criollo goats in Argentina. The high discriminatory value of the withers height, agrees with Devendra and Mcleroy (1982)  using in the classification of tropical goat breeds. Body length  and chest circumference showed significant differences between brown and gray Bengal goats in India (Mukeherjee et al 1979).Contrary to the results of Herrera et al (1996) the importance of head length in discrimination among goat populations was small. Head length showed small discriminating power in native goat breeds of Jordan also. (Zaitoun et al 2005). On doing cluster analysis, it was found that Trivandrum, Kottayam and Kozhikode populations formed a group and Thrissur formed a separate group.

Multivariate analysis of variance made taking three variables, body weight, peak yield and prolificacy, revealed that the goat populations studied were significantly different.  Body weight and peak yield contributed a cumulative 99.9 per cent of total variation.  Low differentiation between these populations was supported by assignment test results. Only about 47.3 per cent of individuals were correctly classified to the population from which they were sampled.  Low assignment rates may indicate either high gene flow or low power to assign because of too few variables used in analysis. (Hirbo et al  2006).  The distance value (D²) calculated between populations revealed that the goat population of Thrissur district was more distant from all other populations. (Table 5)  Cluster analysis using the above three variables indicated that the goat populations of Trivandrum, Kottayam and Kozhikode formed one cluster.  Genetic distances revealed no specific link with geographic distance. Similar pattern was observed by Kotze et al (2004) in Kalahari Red goat populations of southern Africa.

Conclusion

• Local goat populations in Kerala were not very distinct and variations were seen with respect to coat color, presence or absence of horns, tassels or beard and hair pattern.  Based on morphometric measurements, the most discriminative variables were head width and body length, followed by shin circumference and rump length.Trivandrum, Kottayam and Kozhikode populations were grouped together in one cluster based on the distance value.  Thrissur population was distant from all other populations.  Biometrical measures of variation revealed distance values having no specific link with geographical distance.

References

Acharya R M 1982  Sheep and Goat Breeds in India. Animal Production and Health Paper No.30, Food and Agricultural Organization of the United Nations, Rome, pp.190

AICRP Research Report 2005 All India Coordinated research project on “Malabari goat improvement”, Kerala Agricultural University, Thrissur, 45p

Bindu K A 2006 Study of genetic variability of Malabari goats (Capra hircus) utilizing biochemical and immunological markers. PhD thesis submitted to the Department of Animal Breeding and Genetics, Kerala Agricultural University, Thrissur.

Capote J, Delgado J V, Fresno  M, Camacho M E and Molina A 1998 Morphological variability in Canary goat population. Small Ruminant Research.  Volume 27.pp.167-72.

Chatfield C and Collins A J 1980. Introduction to Multivariate Analysis. Chapman and Hall Ltd, London.

Crepaldi P, Negrini R, Milanesi E, Gorni C, Cicogna M and Marsan P A 2001 Diversity in five goat populations of the Lombardy Alps: comparison of estimates obtained from morphometric traits and molecular markers. Journal of Animal Breeding and Genetics. Volume 118. pp. 173-89.

Devendra C and Mcleroy G B 1982  Goat and Sheep Production in the Tropics, Longman, UK, pp.271

Hensen P J 1990 Effects of coat colour on physiological responses to solar radiation in Holsteins. Veterinary Record. Volume 127. pp. 333.

Herrera M, Rodero E, Gutierrez M J, Pena F and Rodero J M 1996 Application of multifactorial discriminant analysis in the morphostructural differentiation of Andalusian caprine breeds. Small Ruminant Research. Volume 22. pp.39-47.

Hirbo J, Muigai A, Naqvi A N, Rege E D and Hanotte O 2006 The genetic diversity of trans-Caucasian native sheep breeds. Asian-Australian Journal of Animal Sciences.Volume19.pp.943-952

Kaura R L 1952 Indian Breeds of Livestock. Prem publishers, Lucknow.

Kotze A, Swart H, Grobler J P and Namaangani A 2004 A genetic profile of the Kalahari red goat breed from southern Africa. South African Journal of Animal Science. Volume 34.pp.10-12.

Lanari M R, Toddeo H, Domingo E, Centeno M P and Gallo L 2003 Phenotypic differentiation of exterior traits in local Criollo goat population in Patagonia(Argentina). Arch. Tierz. Dummerstorf. Volume 46. Article 4.pp.347-356

Livestock Census 2003  Seventeenth Indian Livestock Census Report. Department of Animal Husbandry, Dairying and Fisheries, Ministry of Agriculture, Government of India.

Mukeherjee D K , Singh C S P and Mishra H R 1979 A note on some phenotypic parameters in grey and brown Bengal goats. Indian Journal of Animal Sciences. Volume 49 .pp.671

Raghavan K C, Raja T V and Sasikanth V 2004 Malabari goats. Proceedings of seminar on ‘Goat genome’. pp.101-104.Organised by Central Institute for Research on Goats, Makhdoom on April 5-6, 2004.

Seena T X 2006 Polymorphism of ovine fecundity gene linked microsatellite markers in Malabari goats. MVSc thesis submitted to the Department of Animal Breeding and Genetics, Kerala Agricultural University, Thrissur.

Shanmugasundaram K S 1957 Birth among goats. Indian Veterinary Journal. Volume 34. pp. 107-17.

Zaitoun I S , Tabbaa  M J  and Bdour S 2005 Differentiation of native goat breeds of Jordan on the basis of morphostructural characteristics. Small Ruminant Research.Volume 56 pp.173-182

Received 1 October 2009; Accepted 13 April 2011; Published 1 May 2011

Go to top