Goats have been much appreciated by primitive man mainly because of their small size, easy movement to harvest their food and their docility; However, in recent years, the importance of this domestic species with great productive and reproductive potential has been relegated, without considering that goats offer enormous development prospects, mainly due to their high milk production potential and the organoleptic characteristics of their milk. meat. The good adaptability of goats to slopes, degraded soils and areas with scarce forages, has contributed to the emergence of numerous and small productions, which have allowed the production of goat’s milk in different countries to be increasingly significant (1 ). According to the land use (feeding system), in small ruminants (sheep and goats) there are intensive production systems , semi-intensive (semi-stable) and extensive (2).

There are many factors that modify the production of goat’s milk, which can be interrelated, making it difficult to determine the individual influence that each of them can exert separately. For better understanding, they have been divided into two types; those of a genetic nature and those related to the environment (3). Among those related to the environment, there is food (understood as the quality and quantity of food consumed by goats), this is considered the determining factor on the production of goat milk, as it defines the production system and production volume. There is a high relationship between the amount and composition of the daily diet and the production requirements; Dietary variations can bring important changes in the production and composition of milk (4,5). In Colombia, few research studies have been carried out on the voluntary consumption of dairy goats under different systems and the volume and compositional quality of the milk of mixed-breed goats.

The general objective of this work was to evaluate the effect of the production system on the consumption of food and the production and composition of goat milk in Antioquia.

Study site. Three production systems were evaluated in different agroecological areas of the Department of Antioquia, as follows: system A1 (located in the municipality of Sopetrán, at a geographical latitude of 6.50º and a length of -75.75º, tropical dry forest life zone (6), semi-extensive production system, considered as a natural silvopastoral system, since the animals graze on pastures of natural plant succession, with supplementation in pasture and areas of shelter or artificial shade, system A2 (in the municipality of Barbosa) , latitude 6.28º and longitude -75.45º, pre-montane humid forest life zone (bh-PM) (6), complete stabling with different grasses in the corral, forages and cut grass: King grass (Pennisetum purpureum), Maralfalfa (Pennisetum violaceum), Morera (Morera sp.) And Golden Button (Tithonia diversifolia), in four feeding times); and system A3 (located in the municipality of Guarne, latitude 6.27º and longitude -75.44º, very humid pre-montane forest (6), the animals remain stabled, the food supply is made up of various fodder forages Mainly cut grass. The forage mix is ​​taken to a shredding machine with diesel engine to reduce the size of the particles and to offer in the feeders twice a day (morning and afternoon). The housing is group, the pens are shared by several animals between males and females).

Study animals, information collection and laboratory analysis. A total of 71 goats were evaluated for milk production, at between 80 and 100 postpartum days, in two periods (2010 and 2011). The production of individual milk was weighed on a LEXUS ELECTRONIC SCALES® electronic scale for 5 consecutive days, during each period, according to the milking conditions of each system; a portable ultrasonic milk analyzer(EKOMILK M FAST MODEL®) was used to determine its composition.

For the bromatological composition analysis of the food offered, all the food samples were taken to a MEMMERT forced ventilation oven at 65°C for 72 h and subsequently ground using a stationary mill with a 1 mm sieve (7); the percentage of dry matter (DM) was determined using a moisture balance (O’HAUS®), after a drying process with infrared radiation; the percentage of crude protein (PB) by Kjeldahl method (7), the percentages of fiber in neutral detergent (NDF) and fiber in acid detergent (ADF) by the Van Soest method (8); and gross energy (EB) using an IKA Specials® calorie adiabatic pump. To determine the values ​​of metabolizable energy (ME) in food, the value of the digestibility of organic matter (DOM) was previously estimated, based on the in vitro digestibility of dry matter.

The consumption of individual dry matter was estimated from the fecal production per animal and the digestibility of the food consumed. To find fecal production, chromium oxide was used as an external marker, supplied individually in two fractions of 0.5 g per day and was calculated through mathematical expression (9). The in vitro digestibility of the dry matter was determined by the techniques described by several authors (10-11).

Results Analysis. A multifactorial design was carried out, where the following factors were considered: production system, race, period and the triple interaction between them. The information collected was analyzed through the PROC MIXED procedure and the means found for each treatment were compared using the LSMEANS procedure of the SAS® statistical package (version 9.1). The assumptions of the variance analysis for all the variables under study were validated. The dry matter consumption variable did not present homogeneity in the data, therefore they were transformed and analyzed with the PROC GLM procedure and through non-parametric statistics using the range methodology (12).

Table 1 describes the composition of the forages offered in each production system. It shows that the highest percentage of in vitro digestibility of dry matter, in addition to the highest levels of crude protein (PB) and metabolizable energy (ME) were obtained in system A1, during period one. The highest value of FDN (73.9%) was shown in the forages offered in A2, during period one.

Table 1 Table 1 Table 1. Composition and in vitro digestibility of the dry matter of food

Períod	System	MS	PB	FDN	FDA	EB	DIVMS	EM
1	A1	14.1	13.7	45.1	39.4	3925	73	2,7
	A2	11.9	6.2	73.9	51.1	3699	62.9	2.3
	A3	20	9.5	64.9	40.4	3826	65	2.4
2	A1	29	12.6	58	29.7	4101	67.1	2.5
	A2	24.6	10.1	52.8	26.2	3620	71.7	2.7
	A3	17	9.5	65.6	37.8	3884	66.7	2.5
Dry Matter - MS (% of MS), Gross Protein - PB (% of MS); Neutral detergent fiber FDN (% of MS) Acid detergent fiber FDA (% of MS); Gross Energy - EB (Kcal/Kg MS); In vitro Digestibility of dry matter - DIVMS (%), Metabolizable Energy EM (Mcal /Kg of MS).

In period two compared to period one, the highest percentages of dry matter were found in the food offered in the three production systems, with A1 (29%), versus A2 and A3 (24.6% and 17% of MS, respectively).

Table 2 shows the average values ​​of dry matter consumption - MS, crude protein - PB and fiber in neutral detergent - NDF for each factor (race, period and production system). In the production systems, DM consumption was similar between them, varying between 1042.2 and 1332.9 g/animal/day (p>0.05). Despite the observed variation in the data, in general, DM consumption was not affected by the combined effect of the evaluation period, race and production system.

Table 2. Table 2. Table 2. Consumption of MS, PB and NDF for breeds, periods and production systems

Variables
Effect		MS Consumption (g/animal/day)	PB Consumption (g/animal/day) Dry base	Consumption FDN (g/animal/day) Dry base
Breed	Alpine	1072.6 ± 28.9	107.2 ± 3.0	637.6 ± 16.9
	Saanen	1377.2 ± 19.3	142.1 ± 2.0	800.5 ± 11.3
Períod	1	1316.5 ± 22.8	124 ± 2.4	805.7 ± 13.4
	2	1133.2 ± 24.8	125.4 ± 2.6	632.4 ± 14.6
Production system	A1	1042.2 ± 44.3	147.5 ± 4.6 a	486.2 ± 26 b
	A2	1299.5 ± 22.5	100 ± 2.3 b	804.5 ± 13.2 a
	A3	1332.9 ± 43.9	126.6 ± 4.6 a	866.4 ± 25.7 a
Different Letters represent a significant difference (p<0.05)

The animals in systems A1 and A3 had a higher PB consumption (147.5 and 126.6 g / animal/day, respectively) than those in the A2 system (100 g/animal/day) (p <0.05). NDF consumption among production systems was significantly lower in the A1 (486.2 g / animal/day) compared to the values ​​recorded in the A2 and A3 (804.5 and 866.4 g/animal/day, respectively).

The effects of breed, period and production system on milk production and composition are described in table 3. Production system A1 presented the highest milk production with 1.2 liters/animal/day, superior value and statistically different (p<0.05) from those found for production systems A2 and A3 (0.6 liters/animal/day).

Table 3. Table 3. Table 3. Average and standard deviation of the daily milk production adjusted to 4% of fat, grams of fat, protein and non-fat solids for the factors of breed, period and production systems.

Variables
Effect		Production (L/animal/day)	Fat (g/L)	Proteín (g/L)	Non Fat Solids (g/L)
Breed	Alpine	0.8 ± 9.9	33.6 ± 0.4	22.9 ±0.3	58.6 ±0.8
	Saanen	0.8 ±7.7	37.9 ±0.3	24.1 ± 0.2	63.5 ±0.6
Period	1	0.7 ±9.2b	29.3 ± 0.4 b	15.4 ±0.3 b	45.5 ±0.7 b
	2	1.0 ±8.1a	42.2 ± 0.3 a	31.7 ±0.2 a	76.6 ±0.6 a
Production System	A1	1.2 ± 0.5 a	53.1 ±20.8 a	33.2 ±19.3 a	84.7 ±42.1 a
	A2	0.6 ±0.3 b	27 ±12.7 b	18.5 ±9.1 b	49.5 ±22.6 b
	A3	0.6 ±0.2 b	27.1 ±8.1 b	19 ±8.1 b	49 ±18.6 b
Different letters in each column show significant difference (p<0.05).

Milk production per day and its components were significantly affected (p<0.05) by the triple interaction, observing how the period affects production in each breed and production system.

The effect of the breed, period and production system factors was not statistically significant for the dry matter consumption variable per day, high variability and numerical differences were found in the results, with a minimum value of 458.6 and a maximum value of 3835.4 g/MS/day.

The effect of the production system on the dry matter consumption per day, although it was not significant, in the A3 system showed a higher average consumption (1332 g/MS/day), with an intermediate value in A2 (1299g/MS/day) and the lowest value in A1 (1042g/MS/day), results that may be a consequence of the geographical location, environmental conditions and general management of each production system, such as the supply of food in quantity and quality (different forage species) and the management of the animals (stables and paddocks). The results of the investigation are within the limits reported for dairy goats, minimum 3% of live weight, in goats of high milk production (above 1 kg of milk/day) can reach consumption of 5% of live weight (5); and even, they can consume up to 7% of dry matter (DM) of their live weight, compared with the consumption of 3-4% MS in cows (13).

Variability was found in the composition of the food offered to the goats in the production systems for the two periods evaluated. Crude protein percentages ranged from 6% to 13.7%, Neutral Detergent Fiber-DNF ranged from 45% to 73%, Metabolizable Energy - ME from 2.3 to 2.7 Mcal and in vitro digestibility from 63% to 73%. These results show variability of forage supply in each production system, which can be attributed to the different agroecological conditions in which the production systems and the quality of the soils were found, being the compositional quality of the forages one of the most influential factors in the consumption of food and the productive behavior of goats.

Similar values to those consumptions of dry matter found in this research, were found in cross-bred goats stabled consumptions of 900 g MS/animal/day (13), in Saanen goats in grazing, 1250 g of dry matter per day (14); in Saanen, Toggenburg and La Mancha goats, housed in individual cages, they found consumptions of dry matter of 760 and 1170 g per day (15); and a consumption of 780 g per day in goats under different grazing management conditions (15).

In the production system A1, the highest in vitro digestibility of dry matter of the forages consumed was found (73%), this result suggests a good efficiency of goats and better quality of the food offered, which can have a greater nutrient contribution to milk synthesis; The above can be related to the results of milk production per day, where the A1 system showed higher milk production (1.2 L/day), compared to the A2 and A3 systems (0.65 and 0.63 /day, respectively , being statistically significant, these results can be attributed to factors such as the quality of the diet and its digestibility, in addition to “animal welfare”, when the animals are in a grazing system.

The results found in this investigation are within the reported ranges; the fat component towards the higher limit and above average 5.74% (53.13 g) for the A1, 4.27% (27.8 g) for the A2 and 4.74% (28 g) for the A3, showing a positive relationship between milk quantity and total solids content, since A1 presented the highest production.

Protein in milk was 3.13% (33.57 g) for A1, 3.06% for A2 and 3.29% for A3, which may be an indicator of good nutrient metabolism in goats. Non-fat solids in milk - SNG show a behavior similar to the protein results, considering that it is the main component of this parameter, in addition to mineral salts and lactose.

In conclusion, the production system influenced the production and composition of the milk, with greater production being observed in the silvopastoral production system of natural succession, which can be attributed to a better selection of the forages by the animals (evidenced by the quality of the bromatological composition that was found), this because the animals were grazing, different from the production systems where the animals were in complete stabling.

On the other hand, the production system had no effect on the consumption of dry matter, however, these data were very variable, the values ​​of in vitro digestibility of the dry matter were found to be from medium to high