Body measures, growth curves and body weight prevision of alpacas (Vicugna pacos) reared in Italy

Abstract

Alpacas represent an exotic species for Italy, introduced no more than 20–30 years ago. Nevertheless, they are currently the most widespread camelid species reared in this country, highlighting the growing interest in breeding this species. So, it is useful to assess parameters on their growth in relation to the new environment. Thus, body weight and body measures were recorded on 49 healthy alpacas of Huacaya type (27 females, 22 males), along a period of 27 months. Overall, 111 individual observations were made. Data were analysed by ANOVA, as growth curves were estimated by applying the Gompertz’s model. Multiple regression was finally used to estimate body weight from body measures. Body weight and linear measures, except for the distance between iliac crest and the length of the rump, were influenced by the category (calves, yearlings, adults) and by the age class within category (p < 0.05), while gender and interaction between gender and category were never significant (p > 0.05). Asymptotic weigh of males was 8.25% higher than females (p < 0.05). Moreover, males at maturity had higher withers (+0.96%), longer body (+2.70%) and chest measurements (+3.18%) than females. Asymptotic rump width (both at the ilium and at the ischium) was higher in females (respectively + 6.19%; +12.90%). According to our work, the best equation for the estimate of body weight from body measures was the following: 5.691 + 0.00005624 * chest circumference³ + 0.00002298 * trunk length³ − 1.155 * rump width between ischial tuberosities + 0.00001545 * height at rump³ (R²=0.957; SE of estimate = 4.36 kg), that is important for a better management of the species.

Highlights

• Alpacas are the most widespread camelid species in Italy.

• Rump dimensions develop earlier than other body measures.

• It is possible to improve the management and welfare of animals by estimating the live weight starting from somatic measures.

Keywords:

• Alpaca
• body weight
• body measures
• growth curves
• weight prevision

Introduction

The introduction and breeding of exotic species needs to be carefully managed. Consideration of the species ability to adapt to different climates and nutritional sources, as well as the risk of pathogen exposure to both resident and the introduced species, is needed. The Camelids (Order Artiodactyla, suborder Tylopoda, represented by several species, such as camel, dromedary, llama, alpaca, guanaco, vicuna) represent exotic populations for Europe; however, in Italy, they are bred and are subjected, like all other livestock species, to legal provisions, aimed at their registration in the National Data Bank (BDN) (Ministerial Decree 2 March 2018, attachment 2).

For the European reality the breeding of camelids represents a ‘novelty’, introduced only in the middle ‘90s (Berna 2006; Strozzi 2010; Neubert et al. 2021), while, on the contrary, in South America it is an older tradition. The largest European population is reported as 35,000 alpacas living in Great Britain (Krajewska-Wedzina et al. 2020). Since their domestication, which took place about six thousand years ago, llamas and alpacas have represented, and still represent, essential sources of sustenance for the Andean populations: the alpaca bred to produce fine fibre and meat, and the llama as a multifunctional animal (fibre, meat, transport, skin, and manure) (Frank et al. 2006).

As regards alpaca, more than 96% out of a population of about 3 million heads present in South America live between Peru and Bolivia (Otazù 2005; Ansaloni et al. 2013). The first recording of alpaca breeding in Italy dates back to the end of the ‘90s (Berna 2006). Since then, the herds of camelids, in general, and, in particular, of alpacas, have experienced an exponential increase (Strozzi 2010). Bittante (2011) refers that in DAD-IS of FAO camelids were not registered in Italy, even if the number of alpaca herds has been slowly growing. Currently, based on the data provided by the BDN, the most widespread camelid farms in Italy are those of alpacas (687), followed by llamas (334), camels (64), dromedaries (26), guanacos (5) and vicunas (2). (https://www.vetinfo.it/j6_statistiche/#/, accessed on 12.7.2023). The number of alpacas heads is not reported, but a rough estimate indicates about one thousand (https://www.pubblicitaitalia.com/carne/prodotti/eurocarni/2010/5/9790, accessed on 12.7.2023).

The adaptation of an exotic species can be monitored through the evaluation of the morphological characteristics of the animals and the calculation of their growth curves, also in relation to the values expressed by the animals bred in their areas of origin. To date, there are few research that evaluate the adaptation of the alpaca to Italian environment (Strozzi 2010; Tamburini et al. 2011); likewise, as regards to food rations adopted in our country, they could be far from the real needs of the animals. Alpacas could show a reduced adaptation to environmental conditions through an alteration of their growth, being the result of nutritional deficiencies (e.g. due to the inability to exploit fully forage resources different from the usual ones, palatability and digestive issue, and the adoption of incorrect rations in relation to the needs of the animals) or thermoregulation issues (e.g. the extra use of food energy for the maintenance or dispersion of body heat and not for growth).

The aims of this research were, therefore to, i) evaluate the morphological characteristics of alpacas bred in northern Italy, ii) calculate the growth curves and iii) propose equations for estimating the live weight through somatic measurements.

Materials and methods

Animals and measures

The research was conducted on two farms, located in Northern Italy, Emilia-Romagna region, in the province of Parma and Reggio Emilia. In the first herd about fifty animals are reared and about twenty in the second herd.

The rations were represented by hay from permanent meadows, administered ad libitum, and by a daily supplement (approximately 100 g/d) of the same concentrate, formulated for growing and lactating cattle. The chemical-centesimal composition of the forages and of the concentrate used in both herds is shown in Table 1. The animals were also free to graze on pasture (three hectares in the first, and one hectare in the second herd), which anyhow appeared to be heavily exploited and unable to provide a significant sustenance.

Table 1. Chemical composition (as fed basis) of the forages and concentrate.

		Permanent meadow hay (Herd 1)	Permanent meadow hay (Herd 2)	Concentrate
Moisture	%	8.83	8.81	10.41
Ash	%	6.78	7.27	9.38
Crude protein	%	6.47	6.48	12.82
Ether extract	%	1.56	1.48	3.53
Crude fibre	%	29.99	31.12	10.52
NSC	%	46.37	44.84	53.34
NDF	%	62.34	63.86	31.77
ADF	%	31.89	37.42	12.91
ADL	%	4.70	5.61	3.40
Calcium	g/Kg	5.46	5.21	17.03
Phosphorus	g/Kg	1.36	1.36	6.73
Magnesium	g/Kg	0.92	1.27	2.62
Sodium	g/Kg	0.10	0.50	3.06
Potassium	g/Kg	12.55	14.94	12.66
Iron	mg/Kg	293.44	296.24	452.72
Zinc	mg/Kg	19.84	18.49	143.28
Copper	mg/Kg	3.59	3.94	20.71
Manganese	mg/Kg	37.12	41.35	140.59

NSC non-structural carbohydrates; NDF neutral detergent fiber; ADF acid detergent fiber; ADL acid detergent lignin.

A total of 49 healthy animals of the Huacaya type were used in this study, 27 females and 22 males. Of them, 17 (10 females and 7 males) were measured and weighted only one time, being already adult, whereas the remaining growing animals were submitted to repeated sessions (mean 2.6, range 2–5) during a period of 27 months. The complete dataset contained 111 individual records. The longitudinal measurements were always carried out by the same persons with the same instruments. All measurements were performed on standing animals and represented a routinely operation performed in the farms; no pain was induced to animals that were gently restrained by their owners.

The measurements of body weight (BW) were performed using a portable seesaw (model TE1000L, Tassinari Bilance, Italy) as those of body using a metric string with a millimetre precision or the Lydtin’s measuring stick with a centimetre precision. More precisely, the following measures were recorded:

• height at withers (HW), from the top of the withers to the ground;

• height at rump (HR), from top of rump to ground;

• chest circumference (CC), behind the rear edge of the shoulders at the point of least perimeter;

• flank circumference (FC), in front of the anterior edge of the stifle, at the point of least perimeter;

• body length (BL), from the anterior edge of the shoulder to the tip of buttock;

• trunk length (TL), from the posterior edge of the neck to the base of the tail;

• chest length (CL), from the tip of the shoulder to the posterior margin of the last rib;

• rump width 1 (RW1), distance between iliac crest;

• rump width 2 (RW2), between the ischial tuberosities;

• rump length (RL), from the point of the ilium wing to that of the buttock.

Statistical analysis

The weights and linear measures recorded on the animals were initially subjected to analysis of variance, applying a model including, as fixed factors, the category (3 levels: calves, up to one year; yearlings, from 1 to 2 years; adults, over 2 years), the age-class within the category (7 levels: 0–60 d, 61–120 d, 121–180 d, 181–270 d, 271–365 d for calves, 366–730 d for yearlings, >730 d for adults), the gender (2 levels), the year of birth (11 levels), the interaction between gender and category, and, as a random factor, the sire by year of birth (15 levels). Since the geographical and altimetric conditions and the nutritional management of the two herds were very similar, and, also due to the lower number of records in one of them, it was thus preferred to combine the data of the two herds in a single dataset and not consider in the model the effect of the herd.

The weight and the biometric data were then subjected to non-linear regression analysis, applying the Gompertz’s equation (Sabbioni et al. 2011; KMC Tjørve and E Tjørve 2017), in the following form: [1] $$Y=A\mathrm{ }\text{exp }(-\text{exp }(-b\mathrm{*}(\mathrm{t}-c)\left)\right)$$[1]

Where Y is the BW (kg) or body measure (cm) at day t, A is the mature BW (kg) or body measure (cm), b is the rate of growth, and c is the age (d) of maximum growth. Following KMC Tjørve and E Tjørve (2017), the maximum growth rate was calculated as (b*A)/e, where e is the base of natural logarithm.

Comparisons of the equation parameters A, b and c between sexes were done by means of the Student t-test, using the pooled estimates of standard error to determine significant differences (Pilla 1985). The significance level was stated at p < 0.05.

Body weight and body measures were then submitted to correlation analysis, and to multiple regression analysis, to assess body weight from body measures, with or without the age at the measurement. Both a full model and a reduced one, with independent variables chosen by means of the stepwise procedure were applied. Finally, models built with measurements in the linear, quadratic and cubic form were tested, both with and without age at measurements, and submitted to stepwise regression analysis. The best prediction model was chosen by means of improvement of R² and reduction of standard error of estimate of the dependent variable (SPSS, ver. 28.0.1.0, 2021).

Results and discussion

Weight and somatic measures

Body weight and linear measures (Table 2), except for those relating to RW1 and RL, were influenced by the category and by the age class within the category (p < 0.05); the year of birth and the sire influenced (p < 0.05) a lower number of parameters, whereas gender and its interaction with category were never significant (p > 0.05); the ANOVA model explained a quote of variability ranging between 79.7% (RW2) and 97% (CC). Rump measures showed the lowest coefficients of determination (79.7%–88.3%).

Table 2. Analysis of variance (F values) of weight and body measures in alpacas.

		Category (A)	Age class within category	Gender (B)	Birth year	Interaction (A*B)	Sire	R^2
Degrees of Freedom		2	4	1	10	2	14
BW	kg	195.7**	98.2**	3.1	15.7	51.2	59.0*	0.960
HW	cm	106.8***	125.6***	23.8	27.5**	0.7	23.7	0.941
HR	cm	106.6***	118.1***	2.8	30.9**	10.7	27.5	0.934
CC	cm	181.6***	264.1***	18.3	77.4***	3.2	93.1***	0.970
FC	cm	213.3***	285.3***	41.3	57.3**	18.6	93.6**	0.925
BL	cm	74.2*	154.9***	19.0	47.5**	5.2	115.5***	0.944
TL	cm	218.3***	125.1***	1.00	31.4*	6.8	80.3**	0.934
CL	cm	68.5***	46.9***	22.2	22.7**	11.9	29.2*	0.919
RW1	cm	0.9	8.5	3.1	11.4**	1.0	9.0	0.838
RW2	cm	8.5*	5.8*	1.5	3.1	3.2	2.6	0.797
RL	cm	3.9	10.7*	9.0	12.3**	7.9	7.5	0.883

* : p < 0.05; **: p < 0.01; ***: p < 0.001.

BW Body weight; HW Height at withers; HR Height at rump; CC Chest Circumference; FC Flank circumference; BL Body length; TL Trunk length; CL Chest length; RW1 Rump width 1; RW2 Rump width 2; RL rump length.

The absence of a significant effect of gender and gender*category interaction on BW and body measures confirms the scarce sexual dimorphism shown by the alpacas. This observation has been already highlighted by Bacchi et al. (2010) in guanaco and by Wurzinger et al. (2005) in llama. Also, Grund et al. (2018) found no significant effect of sex on the development of weight and body measures of alpacas. Franklin and Johnson (1994) reported that alpacas are sexually monomorphic, showing a sexual dimorphism only in the canine teeth, with the males having larger teeth for fighting.

The least squares mean (LS means) of body weight and somatic measures in relation to the category and to the age-class within the category are reported in Tables 3 and 4, respectively. All measurements showed smaller values in calves than in the other categories (p < 0.05), except for RW1 and RL. This means that those two latter measurements are able to evaluate growth very early during the life of alpacas; in particular, since the mean age of calves at the measurement was 159 d, we can argue that at about 5–6 months of age the rump has reached a complete development. In sheep, Sabbioni et al. (2016) showed that rump measures were earlier than all other body measures, probably due to the evolutionary need to avoid reproductive problems in early deliveries. In guanaco, Bacchi et al. (2010) did not highlight significant differences in body measurements between calves and yearlings.

Table 3. LS Means (±SE) of body weight and body measures of alpacas as related to age category.

Category		Calves	Yearlings	Adults	RMSE
Age	d	159a ± 83	523b ± 85	1894c ± 89	447
BW	kg	25.04a ± 3.87	38.69b ± 3.56	65.86c ± 3.51	5.59
HW	cm	72.75a ± 2.49	83.38b ± 1.91	87.80b ± 1.32	3.06
HR	cm	74.37a ± 2.53	84.66b ± 1.97	89.79b ± 1.47	3.27
CC	cm	77.93a ± 2.88	92.67b ± 2.21	95.50b ± 1.53	3.54
FC	cm	53.60a ± 3.69	68.12b ± 2.83	75.45c ± 1.96	4.54
BL	cm	67.35a ± 3.16	76.72b ± 2.42	78.78b ± 1.68	3.89
TL	cm	50.09a ± 2.98	63.31b ± 2.28	73.15c ± 1.58	3.66
CL	cm	34.15a ± 2.23	41.38b ± 1.71	47.13c ± 1.19	2.75
RW1	cm	19.31 ± 2.05	19.32 ± 1.79	20.20 ± 0.91	1.89
RW2	cm	8.97a ± 1.44	12.11b ± 1.26	12.23b ± 0.67	1.34
RL	cm	22.31 ± 2.02	24.05 ± 1.76	25.21 ± 0.95	1.92

a,b,c different for p < 0.05.

BW Body weight; HW Height at withers; HR Height at rump; CC Chest Circumference; FC Flank circumference; BL Body length; TL Trunk length; CL Chest length; RW1 Rump width 1; RW2 Rump width 2; RL rump length.

Table 4. LS Means (±SE) of body weight and body measures of alpacas as related to age class within age category.

Category		Calves					Yearlings	Adults	RMSE
Age-class		1	2	3	4	5	6	7
Age	d	24	71	147	239	313	523	1894	447
BW	kg	16.2 (5.0)	21.6 (5.1)	23.9 (5.6)	29.4 (4.1)	34.0 (3.9)	38.7 (3.6)	65.9 (3.5)	5.6
HW	cm	64.2 (3.6)	71.3 (3.7)	72.3 (3.9)	75.1 (2.3)	80.9 (2.2)	83.4 (1.9)	87.8 (1.3)	3.1
HR	cm	65.7 (3.7)	73.1 (3.8)	73.8 (4.1)	77.1 (2.4)	82.2 (2.2)	84.7 (2.0)	89.8 (1.5)	3.3
CC	cm	61.7 (4.1)	72.0 (4.3)	78.0 (4.5)	89.8 (2.7)	88.2 (2.5)	92.7 (2.2)	95.5 (1.5)	3.5
FC	cm	36.7 (5.3)	46.7 (5.5)	53.9 (5.8)	65.8 (3.4)	64.9 (3.2)	68.1 (2.8)	75.5 (2.0)	4.5
BL	cm	58.6 (4.5)	66.7 (4.7)	71.1 (4.9)	67.3 (2.9)	73.1 (2.8)	76.7 (2.4)	78.8 (1.7)	3.9
TL	cm	40.5 (4.3)	47.3 (4.4)	48.5 (4.7)	54.6 (2.7)	59.5 (2.6)	63.3 (2.3)	73.2 (1.6)	3.7
CL	cm	27.9 (3.2)	32.8 (3.3)	31.8 (3.5)	38.1 (2.1)	40.1 (2.0)	41.4 (1.7)	47.1 (1.2)	2.8
RW1	cm	17.5 (2.6)	19.8 (2.6)	19.4 (2.8)	19.1 (2.0)	20.7 (1.9)	19.3 (1.8)	20.2 (0.9)	1.9
RW2	cm	6.8 (1.8)	8.3 (1.9)	8.8 (2.0)	10.0 (1.4)	10.9 (1.3)	12.2 (1.3)	12.1 (0.7)	1.3
RL	cm	19.6 (2.6)	22.3 (2.6)	24.1 (2.8)	23.0 (1.9)	22.8 (1.9)	24.1 (1.8)	25.2 (1.0)	1.9

BW Body weight; HW Height at withers; HR Height at rump; CC Chest Circumference; FC Flank circumference; BL Body length; TL Trunk length; CL Chest length; RW1 Rump width 1; RW2 Rump width 2; RL rump length.

In the comparison between yearlings and adults, the latter ones showed higher values (p < 0.05) for BW, FC, TL, and CL. Therefore, it seems that already between the first and second year of age there is a slowdown in growth in the alpaca bred in Italy in terms of height, CC, BL, and RW2.

The weight data reported in the present study were then compared with those recorded by Davis et al. (1991) on Chilean alpacas, imported and bred in New Zealand. The weights of the adult animals and of the crias at about 4 months of age were very similar from the comparison of the two studies, while they were lower for animals of 5 and 10 months in the Italian population. The weights of Italian crias at these ages were however higher than those of animals reared in Peru (Cristofanelli et al. 2004). McGregor (2006) reported for adult Huacayas a weight range from 77.5 kg (females) to 80.8 kg (males). Body measures were compared with those reported by Tamburini et al. (2011) on alpacas of the age of 4.1 ± 3.2 years: our Italian data obtained on adults at an average age of 5.2 years resulted higher for HW (87.8 cm vs 85.6) and CC (95.5 cm vs 93) and lower for RW (20.2 cm vs 24).

Growth curves of weight and somatic measures

The parameters of Gompertz’s equation for live weight and the main linear measures are shown in Table 5 (the relative graphs are reported in Supplementary Material). While no significant difference concerned the parameter b (p > 0.05), the parameters A and c relating to weight and some linear measures were significantly different in the two sexes (p < 0.05). Asymptotic weight of males was 8.25% higher than that of females (p < 0.05). Moreover, males at maturity resulted higher at withers (+0.96%) and with longer body (+2.70%) and chest (+3.18%) than females. Asymptotic rump width (both at the ilium and at the ischium) was higher in females (RW1 + 6.19%; RW2 + 12.90%), probably due to the need to avoid dystocic deliveries (Sabbioni et al. 2016). Interestingly, days at maximum growth rate (c parameter of the Gompertz equation) were positive for BW and negative for body measures, indicating that alpacas at birth are well developed in size, as body mass develops later. This confirms the findings of Wurzinger et al. (2005) in llamas and Strozzi (2010) in alpacas; by applying the Brody’s growth model, this latest authoress highlights that at birth the weight of alpacas was 11%–12% of adult weight, as measures ranged between 44% (RL in females) and 66% (both HW and HR in females) of asymptotic measure. Moreover, she found that 90% of adult weight was reached around 50 months of age, as 90% of body measures were reached between 9 (HW and HR) and 27 (FC) months of age. Wurzinger et al. (2005) found similar values in llamas (7%–15% of mature weight at birth; 44%–63% of mature body measures – respectively, FC and HW – at birth).

Table 5. Parameters of the Gompertz equation (A, b, c) and standard errors (SE) for body weight and body measures in alpaca (see the related graphics as supplemental material).

	A	SE	b	SE	c	SE	R^2	RMSE
BW
All	69.08	1.60	0.002454	0.000213	241.97	16.96	0.915	6.0
Males	71.37b	2.02	0.002485	0.000304	254.27b	23.47	0.925	5.8
Females	65.93a	2.45	0.002501	0.000321	225.28a	24.63	0.907	6.1
HW
All	89.45	0.65	0.004325	0.000369	−196.45	24.52	0.846	3.86
Males	89.82b	0.78	0.004500	0.000505	−179.42a	33.86	0.848	3.45
Females	88.96a	1.05	0.004264	0.000558	−205.68b	36.33	0.838	4.23
HR
All	91.62	0.67	0.004210	0.000364	−209.13	25.68	0.848	3.85
Males	91.44	0.80	0.004710	0.000557	−177.76a	34.72	0.833	3.58
Females	91.79	1.13	0.003864	0.000499	−233.08b	34.72	0.851	4.12
CC
All	97.77	0.97	0.002889	0.000209	−141.36	21.96	0.891	5.18
Males	97.89	1.38	0.002730	0.000308	−158.48b	40.51	0.881	4.99
Females	97.76	1.41	0.002999	0.000297	−131.66a	26.85	0.896	5.43
FC
All	72.00	1.08	0.003077	0.000324	−121.26	29.57	0.792	5.69
Males	71.89	1.53	0.002765	0.000471	−152.73b	60.32	0.763	5.59
Females	72.29	1.56	0.003307	0.000464	−102.90a	33.22	0.812	5.84
BL
All	83.78	0.81	0.003690	0.000268	−107.90	17.99	0.883	4.52
Males	84.85b	1.01	0.003474	0.000341	−109.50b	29.01	0.893	4.03
Females	82.62a	1.27	0.003939	0.000436	−105.15a	23.79	0.874	4.93
TL
All	74.75	0.83	0.003249	0.000270	−136.81	23.46	0.857	4.43
Males	74.91	1.12	0.003136	0.000393	−138.61b	41.23	0.844	4.33
Females	74.60	1.25	0.003373	0.000394	−132.34a	29.25	0.862	4.60
CL
All	46.69	0.56	0.003556	0.000333	−121.18	24.44	0.823	3.16
Males	47.39b	0.69	0.003417	0.000422	−119.78a	37.34	0.847	2.74
Females	45.93a	0.89	0.003735	0.000537	−121.40b	33.51	0.802	3.50
RW1
All	20.63	0.36	0.004172	0.000635	−126.64	36.65	0.635	2.07
Males	20.03a	0.44	0.004991	0.001166	−112.55a	55.44	0.570	1.90
Females	21.27b	0.59	0.003650	0.000739	−132.70b	48.89	0.676	2.21
RW2
All	10.75	0.27	0.005611	0.001452	−102.66	48.39	0.398	1.63
Males	10.08a	0.29	0.009208	0.004696	−62.27a	63.06	0.290	1.48
Females	11.38b	0.42	0.004693	0.001403	−102.11b	57.54	0.499	1.69
RL
All	25.01	0.39	0.003851	0.000441	−89.81	26.63	0.771	2.11
Males	24.99	0.52	0.003473	0.000625	−134.32b	54.70	0.746	1.95
Females	25.09	0.59	0.003986	0.000630	−73.02a	31.71	0.782	2.27

a,b different for p < 0.05.

BW Body weight; HW Height at withers; HR Height at rump; CC Chest Circumference; FC Flank circumference; BL Body length; TL Trunk length; CL Chest length; RW1 Rump width 1; RW2 Rump width 2; RL rump length.

Also, Grund et al. (2018) found that BW of alpacas did not reach the mature weight at the age of 3 years, ranging at that age between 40 (3rd percentile) and 70 (97th percentile) kg. Our data showed an asymptotic BW near these last value (71.37 kg in males and 65.93 kg in females), reached between 4 and 5 years of age; compared to data reported by Grund et al. (2018), HW in our study resulted higher (89.45 cm in males and 88.96 cm in females vs 87.8 cm), so as HR (91.44 cm in males; 91.79 cm in females vs 89.8 cm), while TL was not comparable because of the different site of the measure.

At the moment of maximum growth (parameter c of the Gompertz model), daily weight gain was 61–65 g/d, respectively in males and females, as growth in size ranged between 0.02 cm/d for RW2 and 0.14 cm/d, both for HW and HR. Castellaro et al. (1998) reported daily weight gains different among seasons in alpacas reared in Chile: 100–200 g/d in spring, 50–100 g/d in winter and negative gains in end of summer and fall. Newborn alpacas grew faster in the first 90 days (110–150 g/d), reaching 75 g/d at 8–9 months. At 3 years of age a residual growth was recorded (10–20 g/d). Other research report data slightly different to those in the present research. Valenzuela et al. (2021) report that daily body weight gain during the first 12 weeks of life of alpacas in the Peruvian Andes ranges between 130 and 170 g/d, in relation to the kind of management system. Brown and Van Saun (2014) report that the growth of alpacas reared in USA reaches plateau around 25 months (65–75 kg) and that the maximum daily gain is observed in the second month of life.

Body weight prediction from body measures

The application of an equation to estimate the BW starting from linear measures is of good use in practice, as the linear measures taken on the animal are much less stressful than weighing, given the temperament of the alpaca which is particularly not inclined to be manipulated by humans.

In our work, all linear measurements were highly correlated with live weight and age at the measurement (Figure 1); in particular, CC showed the highest correlation with BW (r = 0.940; p < 0.01), as found also by Kiesling (1995) on llamas and Grund et al. (2018) on alpacas. Also in other species, such as sheep (Sarti et al. 2003; Sabbioni et al. 2020), cattle (Ozkaya and Bozkurt 2009) and goats (Lan et al. 2021) the CC resulted highly correlated to BW. For donkeys an equation for predicting live weight from body measures (Cappai et al. 2013) included CC besides HW, resulting in high R² and high significance of the independent variables.

Figure 1. Heatmap of correlation coefficients among age at measurement, body weight and linear body measures in alpaca (BW body weight; HW Height at withers; HR Height at rump; CC Chest circumference; FC flank circumference; BL body length; TL trunk length; CL Chest length; RW1 rump width 1; RW2 rump width 2; RL rump length). All correlation coefficients are significant at p < 0.01.

Single body measurements or combinations of these could therefore be predictive of the body weight of the animal. As stated by Riek and Gerken (2007) on llamas, given that the addition of several body measurements adds little to the already good result obtained with a single measurement, it would be convenient to use a single value, in particular the CC, that was confirmed to be the one with the highest degree of correlation with BW. Strozzi (2010) proposed a quadratic equation for the estimation of the live weight of alpacas starting from the measurement of the CC: [(BW (kg) = 0.0162 * CC² − 1.252 * CC + 32,752 (R² = 0.927; SE of the estimate = 4.5 kg)].

In Table 6 some results of the multiple regression analysis are reported. Models with all body measures gave a better prediction than those with body measures chosen by the stepwise procedure. Moreover, models also including age at measurement as independent variable gave a better estimate of BW than those without it. However, according to Grund et al. (2018), the best equation was obtained by including CC, TL and HR at the cubic power (besides RW2 in the linear form, as age was not included in the model by the stepwise procedure). The cited authors found too that BW as a function of the different body measurements follows the model of a cubic function, because of the three-dimensional shape of the body. Moreover, in studies related to allometry (Sabbioni et al. 2016) conducted on sheep, isometry between body weight and linear measures occurred at a value of 0.333 rather than 1. Brown and Van Saun (2014) proposed an equation with a high R² (0.97) for body weight prevision, that included as independent variable only the age until the power of 4.

Table 6. Improvement of accuracy and precision of body weight estimate from body measures with and without age at measurement.

Model	Independent variables	coefficient	R^2	SE
All body measures (cm) without age	Intercept HW HR CC FC BL TL CL RW1 RW2 RL	−44.109 −0.854 0.491 1.335 −0.415 −0.141 0.508 0.405 0.873 −1.672 −0.0239	0.922	6.18
All body measures (cm) with age (d)	Intercept Age HW HR CC FC BL TL CL RW1 RW2 RL	−44.646 0.005 −0.898 0.905 1.025 −0.344 −0.123 0.457 0.140 0.885 −1.288 −0.495	0.935	5.67
Body measures (cm) chosen by stepwise procedure without age	Intercept CC TL RW2	−51.719 1.087 0.382 −1.805	0.912	6.23
Body measures (cm) chosen by stepwise procedure with age (d)	Intercept Age CC TL RW2	−41.500 0.004 0.836 0.398 −1.288	0.926	5.73
Body measures (cm) up to cubic power chosen by stepwise procedure	Intercept CC^3 TL^3 RW2 HR^3	5.691 0.00005624 0.00002298 −1.155 0.00001545	0.957	4.36

HW Height at withers; HR Height at rump; CC Chest Circumference; FC Flank circumference; BL Body length; TL Trunk length; CL Chest length; RW1 Rump width 1; RW2 Rump width 2; RL rump length.

Conclusions

Among the camelids, alpacas have assumed the role of the most farmed species in Italy. Therefore, any study about the adaptation of this species to the Italian environmental conditions must be considered of fundamental importance, aiming to a better animal management and animal welfare.

It is necessary for the future to conduct further studies to evaluate the adaptation of animals to other environments as well, and to give a stronger economic impact to this type of farming.

Ethical approval

All research reported in this research has been conducted in an ethical and responsible manner.

Geolocation information

Varano dè Melegari: 44.69445367257062, 10.002542190167581. Ceredolo dei Coppi 44.55820799216058, 10.449438117968214.

Disclosure statement

The authors report there are no competing interests to declare.

Data availability statement

The data presented in this study are available on request from the corresponding author upon reasonable request.