Morphological measurements and indexes as a tool to support molecular genetic studies: an example in Canary Islands

Abstract

In order to complement previous DNA studies, 81 domestic goats, belonging to the Majorera Breed (MB, 105), Semi Wild Goats (SWG, 31) and Tinerfeña Breed (TF 61), were used. Both populations (MB and SWG), allocated in the same geographical area, were differentiated previously by microsatellites analyses, and TF is from another island. Thirteen morphological measurements were recorded, zoometric indexes were calculated and profile and coat colour were evaluated. For all morphological measurements, SWG were significantly smaller than MB and TF. These results agree with previous data obtained with genetic studies, which show difference between MB and SWG. The results of the zoometric indexes showed that SWG were lighter and shorter in relation to the thoracic circumference and height, suggesting long-term environmental adaptation. In general, higher correlations were observed in SWG than in MB followed by the TF, for example between height at the end of neck and height to the rump (0.880, 0.740 and 0.683, respectively), probably indicating a superior body balance to adapt to ultra-extensive conditions. With regard to other parameters, most SWG presented a subconcave profile (66.7%) compared to TF, which had 71.19% of rectum profile and 40.60% of MB was subconvex. SWG had less complex colour distribution.

Keywords:

• goat
• morphology
• index
• genetic
• characterization

Introduction

Usually molecular genetic research studies have been done on animals that previously have been described morphologically (Capote et al. 1998; Yakubu et al. 2010, no extensive molecular genetic characterisation of Nigerian goats has been published as we have just submitted a paper to that effect), although recently DNA genetic marker studies have been developed on goat population without previous morphological characterisation (Martínez et al. 2006). Goat breeds on Canary Island domestic goats were morphologically described (Capote et al. 1998) and, later, microsatellites and mitochondrial analysis were done (Amills et al. 2004; Martínez et al. 2006). As a result of these previous studies, three different breeds were characterised: Majorera, Palmera and Tinerfeña (Capote et al. 1998) also taking into account productive aspects (Argüello et al. 1999). Later, genetic studies based on blood polymorphism and DNA confirmed the results found using morphological traits, although all breeds had the same origin and a strong phylogeographic relationship (Amills et al. 2004). More recently, Martínez et al. (2006) studied a wider population, including semi wild goats, considered as part of the Majorera Breed (MB) because both share a common habitat in the Fuerteventura island fields (Capote 1989). They confirmed previous results about domestic goats and also showed that one of the semi wild populations did not differ from the MB though, goats belonging to another semi wild population, called Ajui by those authors, did not cluster with the domestic ones, suggesting that they could belong to a different genotype. As historic reports show, the ancient Canary Island humans had two kind of goats, ‘jaira’ in herds managed by shepherds and ‘guanil’, wild or semi wild goats (Tejera and Capote 2005). Recently it has been reported that Ajui population could be an old genotype (Martínez et al. 2006) and, therefore, related to ‘guanil’ goats. All studies done using genetic markers gave important information but nothing has been reported about morphological breed characteristics and the possible relationship with MB, its geographical neighbour. The aim of this study was to complement the DNA studies checking the suitability of the morphological characterisation as a tool to support genetic results.

Materials and methods

One hundred and twelve adult goats consisting of 81 MB (BOC N° 105) reared on 5 different farms, 61 Tinerfeña Breed (North Variety) (TB, BOC N° 102) reared on 4 different farms and 31 semi wild goats (SWG) from the west of the island of Fuerteventura (Spain). The environmental conditions of this area are very arduous, with around 100 mm per year of rainfall, high isolation rate and little vegetation. The semi wild animals are usually rounded up (between 500 and 1000 goats) once or twice per year. This is done to identify them, castrate the males and separate the males that were castrated the previous year, to slaughter for meat. The management usually only takes 2–3 hours, to avoid a additional dangerous level of stress in goats that are frequently already under sub-optimal nutritional conditions. This implies that the farmers have to work fast and be coordinated.

The animals were weighed (W) and 12 morphological measurements were recorded (Figure 1): body length (Bl), breast width (Bw), height at the end of neck (Hn), height to the rump (Hc), height to the breast (Hb), distance between eyes (Dbe), head length (Hl), rump width (Rw), rump length (Rl), thoracic circumference (Tc) and shank circumference (Sc); the thorax depth (Td) was calculated as the height at end of neck minus height to the breast. Using the data, different zoometric indexes were calculated (Aparicio 1960; Sotillo and Serrano 1985). All animals were photographed from both sides of the body and, furthermore, the frontal profile and coat colour were evaluated. To get successful measurements in a short period of time, a group of students from the Veterinary Faculty of Universidad de Las Palmas de Gran Canaria were previously trained and coordinated by several researchers.

Figure 1.  Morphologic measurements recorded in this experiment.

Note: Bl, body length; Bw, breast width; Dbe, distance between eyes; Hn, height at the end of neck; Hl, head length; Hb, height to the breast; Hc, height to the rump; Sc, shank circumference; Tc, thoracic circumference; Rl, rump length; Rw, rump width.

SPSS 17.0 One-factor ANOVA for the body measurements and indexes, Pearson correlation coefficient between measurements and principal component analyses were carried out using SPSS 17.0.

Results and discussion

Table 1 shows the results of morphological parameters measured in MB, TF and SWG. As was expected, all parameters were significantly higher in the MB and TF than SWG (P>0.001). Except that TF had similar Dbe that SWG and MB had their head widest than both. Those differences were probably due to arduous environmental conditions in the grasslands. These grasslands have a low forage availability which could negatively affect animal growth. Previously Mello and Schmidt (2008) reported a decrease in the body measurements of Anglo Nubian goats born in Brazil, as a result of adaptation to lower nutrient intake. Between MB and TF, MB had bigger Hn and Hb; however, both had the same Td, so MB had longer legs.

Table 1. Morphological parameters of Majorera Breed (MB), Tinerfeña Breed (TF) and semi wild goat (SWG).

	MB	TF	SWG		Capote et al. (1998)	Capote et al. (1998)
Measurements	Mean±SD	Mean±SD	Mean±SD	P	MB^a	TF
Height at the end of neck	70.94±3.93^a	69.56±2.38^a	54.42±4.97^b	**	70.70	68.5
Height to the rump	72.91±3.52^a	69.86±2.50^b	55.55±4.39^c	**	73.40	72.2
Height to the breast	38.61±3.04^a	37.30±2.61^b	29.84±2.28^c	**	–	–
Body length	79.55±0.60^a	75.68±4.16^b	60.72±4.77^c	**	70.50	72.4
Breast width	17.20±1.94^a	17.08±2.08^a	12.97±1.58^b	**	20.70	20.9
Thorax depth	31.93±5.19^a	32.25±3.13^a	24.58±4.16^b	**	32.20	32.4
Thoracic circumference	87.68±7.14^a	85.66±6.13^a	68.03±5.99^b	**	96.50	97.1
Shank circumference	9.31±0.71^a	8.95±0.53^b	7.45±0.51^c	**	9.40	9.4
Head length	23.18±2.13^a	23.20±2.56^a	18.42±1.26^b	**	22.00	22.6
Head width	12.96±1.20^a	11.16±1.34^b	10.77±0.62^b	**	13.30	12.8
Rump length	24.20±3.04^a	23.85±2.81^a	17.9±1,66^b	**	22.40	22.5
Rump width	17.33±1.20^a	16.45±1.89^b	12.65±1.36^c	**	16.70	16.9
Weight	55.62±12.18^a	56.10±9.00^a	23.52±5.07^b	**	–	–
Note: Length measurements are in centimetre and weight in kilograms Different letters (a, b, c) in the same row, mean statistical difference (P<0.01).
^aMajorera Breed (raised in Fuerteventura Island) was used in Capote et al. (1998).
**Means with significantly different (P<0.001).

Our results for the MB and TF in the present study agree with those reported by Capote et al. (1998) although these authors found a smaller body length (Table 1). Although there are few references in the literature on the body weight of the MB; nevertheless Argüello et al. (1999) similar body weight of 57.88 kg in the same breed.

Table 2 shows the results of the zoometric indexes. SWG and TF had a lower body index value (Bl/Tc) which means that with the same thoracic circumference the SWG and TF have shorter body lengths. So, MB was more longilinear than the others. In the same way, the relative shortness index (Hn/Bl) showed that MB was longer than SWG and TF. On the other hand, indexes related to the body weight showed that SWG were lighter, in relation to the height, than MB and TF and that they support less weight in relation to shank circumference. These results were probably due to long-term environmental adaptation, which supports that SWG could be an ancestral population which agrees with the findings of Martínez et al. (2006).

Table 2. Zoometric indexes of Majorera Breed (MB), Tinerfeña breed (TF) and semi wild goat (SWG).

	MB	TF	SWG
Indexes	Mean±SD	Mean±SD	Mean±SD	P
Body index	91.6±6.52^a	88.60±5.46^b	87.38±5.69^b	**
Thoracic index	53.75±7.26	53.15±6.16	54.4±11.4	0.806
Cranial index	56.18±5.51^a	48.44±6.43^a	58.61±2.97^b	**
Rump index	71.54±9.59	69.71±10.43	70.75±5.63	0.511
Relative thoracic deep index	45.51±3.98	46.34±3.78	44.9±4.7	0.239
Tranver rump index	24.46±2.07	23.67±2.96	23.32±2.36	0.068
Long rump index	34.5±3.92	34.31±4.02	33.05±3.23	0.194
Relative weight index	78.13±15.05^a	80.57±11.98^a	43.03±7.22^b	**
Relative shortness index	88.85±4.58^a	92.14±5.13^b	92.45±4.42^b	**
Shank weight index	17.35±3.16^a	16.29±2.33^a	32.83±5.79^b	**
Different letters (a, b) in the same row, mean statistical difference (P<0.01). **Measurements with significantly different (P<0.01).

The correlations between zoometric parameters (Tables 3 and 4) showed the importance of the height to the end of neck, as is widely correlated with most of the other parameters in both groups, as occurred with height of the rump, body length and body weight. Ribeiro et al. (2004) also described a high correlation between the height to the end of neck and body weight. As Reis et al. (2004) observed in dairy cows, thoracic circumference was correlated with all studied parameters, except with the height to the breast. It is important because thoracic circumference and height to the end of neck are indicators of the body weight and carcass yield (Araújo et al. 2004). Different authors (Ferreira et al. 2000; Santana et al. 2001) showed that body weight is correlated with all morphological measurements except with height to the breast. In general, the correlations in SWG were higher than in MB and these higher than TF, probably due to the fact that SWG lives on a wild, but homogeneous system and it could indicate a greater body balance to adapt to ultra-extensive conditions.

Table 3. Correlations between morphologic measurements of Majorera Breed (MB), Tinerfeña Breed (TF) and semi wild goat (SWG).

		Hc	Hb	Bl	Bw	Td	Tc	Sc	Hl	Dbe	Rl	Rw	W
Hn	SWG	0.880**	0.556**	0.788**	0.350	0.890**	0.494**	0.386*	0.408*	0.293	0.441*	0.508**	0.616**
	MB	0.740**	0.440**	0.784**	0.381**	0.687**	0.482**	0.406**	0.334**	0.265*	0.386**	0.378**	0.593**
	TF	0.683**	0.217	0.240	0.189	0.578**	0.377**	0.215	0.338**	0.330**	0.147	0.069	0.434**
Hc	SWG		0.545**	0.746**	0.411*	0.752**	0.397*	0.305	0.434*	0.392*	0.447*	0.611**	0.548**
	MB		0.460**	0.686**	0.442**	0.397**	0.466**	0.468**	0.287**	0.230*	0.305**	0.374**	0.586**
	TF		0.328**	0.159	−0.069	0.245	0.361**	0.226	0.332**	0.347**	0.234	0.080	0.395**
Hb	SWG			0.548*	0.331	0.116	0.060	0.210	0.024	0.044	0.392*	0.250	0.362*
	MB			0.140	0.142	−0.350**	0.082	0.195	0.037	0.044	0.299**	0.197	0.169
	TF			−0.193	−0.401**	−0.671**	−0.283*	−0.146	0.132	−0.176	−0.367**	−0.013	−0.316*
Bl	SWG				0.666**	0.598*	0.622*	0.241	0.367	0.335	0.626*	0.531*	0.707**
	MB				0.538**	0.645**	0.632**	0.542**	0.296*	0.338*	0.309*	0.428**	0.681**
	TF				0.421**	0.343**	0.544**	0.223	0.370**	0.222	0.368**	0.260*	0.613**
Bw	SWG					0.236	0.501**	0.248	0.225	0.505**	0.532**	0.695**	0.568**
	MB					0.257*	0.703**	0.548**	0.392**	0.307**	0.309**	0.560**	0.747**
	TF					0.479**	0.536**	0.385	0.192	0.152	0.201	0.332**	0.555**
Td	SWG						0.557**	0.346	0.474**	0.325	0.312	0.469**	0.538**
	MB						0.367**	0.213	0.273*	0.175	0.007	0.189	0.366**
	TF						0.522**	0.285	0.146	0.397**	0.418**	0.064	0.593**
Note: Hn, height at the end of neck; Hc, height to the rump; Hb, height to the breast; Bl, body length, Bw, breast width, Td, the torax depth; Tc, thoracic circumference; Sc, shank circumference; Hl, head length; Dbe, distance between eyes; Rw, rump width; Rl, rump length; W, weight.
*Measurements with significantly different (P<0.05).
**Measurements with significantly different (P<0.01).

Table 4. Correlations between morphologic measurements of Majorera Breed (MB), Tinerfeña Breed (TF) and semi wild goat (SWG).

		Hc	Hb	Bl	Bw	Td	Tc	Sc	Hl	Dbe	Rl	Rw	W
Tc	SWG							0.594**	0.639**	0.687**	0.591**	0.494**	0.852**
	MB							0.725**	0.420**	0.352**	0.284*	0.604**	0.883**
	TF							0.484**	0.568**	0.391**	0.549**	0.392**	0.868**
Sc	SWG								0.609**	0.605**	0.689**	0.339	0.653**
	MB								0.217	0.244*	0.285**	0.622**	0.709**
	TF								0.387**	0.230	0.242	0.323*	0.525**
Hl	SWG									0.684**	0.498**	0.403*	0.576**
	MB									0.307**	0.210	0.368**	0.429**
	TF									0.261*	0.394**	0.226	0.562**
Dbe	SWG										0.694**	0.658**	0.635**
	MB										0.260	0.314**	0.423**
	TF										0.292*	0.157	0.351**
Rl	SWG											0.695**	0.679**
	MB											0.313**	0.379**
	TF											0.154	0.574**
Rw	SWG												0.595**
	MB												0.694**
	TF												0.431**
Note: Hn, height at the end of neck; Hc, height to the rump; Hb, height to the breast; Bl, body length; Bw, breast width; Td, the thorax depth; Tc, thoracic circumference; Sc, shank circumference; Hl, head length; Dbe, distance between eyes; Rw, rump width; Rl, rump length; W, weight.
*Measurements with significantly different (P<0.05).
**Measurements with significantly different (P<0.01).

Figure 2 shows head side view and coat colour percentages. Head side view was different between both groups of goats, thus a higher percentage of SWG were subconcave (66.70%) while 44.60% of MB showed a subconvex and 71.19% of TF had rectum head side view. This parameter was, after DNA, one of the most important, with regard to livestock breeds classification (Aparicio 1960), so these results lend additional support to DNA studies. Otherwise, the results found in MB and TF were similar to those reported by Capote et al. (1998) in the same breed. With regard to coat colour distribution, the TF coat colour was not recorded because all had dark coat as is recorded in the breed standard. SWG showed a higher percentage of black and white animals than MB (74.20% vs. 37.00%, respectively), with MB having a higher percentage of goats with mixed colour (black, red and white). It could be due to the aim of the farmers to easily identify animals that are usually wide spread on poor grasslands.

Figure 2.  Head side view and coat colour (%) of the Majorera Breed, Tinerfeña breed* and semi wild goat.

*The coat colour of Tinerfeña breed is not represented, because all goats are black as it is recorded in the standard breed.

Finally, Figure 3 shows the result of principal component analysis. The animal of the same groups (SWG, MB and TB) cluster together, but the SWG animals are separated of the other, while the MB and TB don't show clear differentiation. In dimension 1, the variable with a higher entity was the weight (0.982) followed by the height at end of neck (0.965) and height to the rump (0.962). So it was clear that the weight and height distance the SWG group of the others groups. While in dimension 2, the most important variable was the body length (0.802) and head width (0.657). The head width can explicate the separation between these groups in the graphic. Finally with this analysis, it was possible to see that the SWG was separated from the other groups according to their morphology measure.

Figure 3.  Principal component analyses of Majorera Breed (1) vs. semi wild goat (2) vs. Tinerfeña breed (3).

Conclusion

The results support previous findings using genetic analysis and, furthermore, establish a deeper divide between both groups, suggesting that both populations, SWG and MB, belong to different breeds. Therefore, morphological measurements could be a tool to complement molecular genetic studies.

Acknowledgements

Authors thank the students of the Veterinary Faculty of Universidad de Las Palmas de Gran Canaria, the Fuerteventura and Tenerife farmers, the farm staff of the ICIA Goat farm, especially Morin and Emma Ball, for reviewing the English grammar of the paper.