Supplements of lactating meat goat does grazing grass/forb pastures

Abstract

Lactating Boer does grazing grass/forb pastures were used in a 16-wk study starting 22 days after birth. Treatments were no supplementation (CO), access to a 20% crude protein supplement block (SB), and placement in a supplement pasture with mimosa (Albizia julibrissin) trees for 6 h 1 day/wk (1X) or twice weekly for 3 h/day (2X). Available forage dry matter in nonsupplement pastures ranged between 2423 and 3477 kg/ha. Treatment did not affect doe average daily gain (ADG), although kid ADG in the first 12 wk differed (P<0.05) between type of supplement and frequency of supplement pasture access (121, 111, 120, and 134 g for CO, SB, 1X, and 2X, respectively). Lactating Spanish does were used in a 12-wk study starting 66 days after kidding. The same CO and SB treatments were employed, but access to supplement pastures was for 24 h 1 day/wk (1X) or 2 days for 6 h/day (2X). Forage dry matter ranged only from 750 to 1530 kg/ha; thus, 0.6 kg/day (as feed) per doe of grass hay was fed after 4 wk. Kid ADG in wk 1–8 was not affected by treatment. Doe ADG was affected by supplementation (P <0.05) and supplement type (P<0.09) (−44, −33, −23, and −12 g for CO, SB, 1X, and 2X, respectively). In conclusion, the use of SB was not beneficial, and infrequent access to supplement pastures had relatively small effects on ADG, perhaps because forage availability and nutritive value were not severely limiting.

Keywords:

• goats
• grazing
• supplementation
• protein bank
• multi-nutrient block
• tree legume

1. Introduction

Infrequent supplementation of cattle and sheep consuming low-protein forage with feedstuffs high in protein, such as once or twice weekly, can improve performance to similar extents as more frequent supplementation (McCollum and Horn 1990; Huston et al. 1999a, 1999b). Likewise, Joemat et al. (2003) did not observe differences in performance by confined Angora does consuming a basal diet of grass hay, 5% in crude protein, supplemented with soybean meal every 1, 4, or 8 days while in late gestation, although in early lactation supplementation every 8 days resulted in greater body weight loss compared with more frequent supplementation. Moreover, Calhoun et al. (1988) noted similar mohair fleece production and fiber diameter for male Angora goats consuming low-quality forage supplemented with cottonseed meal every 1, 2, 3, 4, or 5 days.

Pasture inclusion of leguminous forage offers numerous benefits in ruminant production systems, many of which are also true for leguminous trees. However, the establishment and maintenance of leguminous plants throughout large pasture areas can be challenging. This has contributed to interest in allowing ruminants short periods of access to small areas with high densities or pure stands of legumes, sometimes referred to as ‘protein banks’ (Paladines 1984; Mohamed-Saleem et al. 1986; Moron-Fuenmayor and Clavero 1999). Periodic access of ruminants to pastures high in legumes may provide a more practical or feasible means in some regions of providing supplemental nutrients compared with daily access for short periods of time, such as 2–2.5 h (Paladines 1984; Mohamed-Saleem et al. 1986; Moron-Fuenmayor and Clavero 1999), as well as continual availability of relatively expensive multi-nutrient blocks. Therefore, the objectives of these experiments were to determine the effects of different frequencies and lengths of access to pasture with the tree legume mimosa (Albizia julibrissin Durazz) on performance by lactating meat goat and their kids grazing grass/forb pastures in the summer.

2. Materials and methods

2.1. Experiment 1

A protocol for each experiment was approved by the Langston University Animal Care and Use Committee. Experiments were conducted at the American Institute for Goat Research of Langston University (latitude 35°56′35″N; longitude 97°16′52″W; elevation 261 m) in 10 0.4-ha grass/forb pastures. Thirty-two multiparous Boer does were used with 16 nursing one kid and others two kids. The experiment was 16 wk long and began on 3 June 2008, at which time kids averaged 22 days of age (S.E.M.=2.0); kids were weaned after 12 wk. Does were allocated to eight groups based on litter size, body weight (BW), and kid age. Groups were randomly assigned to four treatments, with two groups per treatment. At the beginning of the experiment, does were orally treated for internal parasites with 0.75 mg/kg BW of moxidectin (Cydectin®; Fort Dodge Animal Health, Fort Dodge, IA, USA) and 14 mg/kg BW levamisole (Leavasole®; Schering-Plough Animal Health, Union, NJ, USA). After 4 wk, the kids were vaccinated for Clostridium perfringens types C and D and tetanus (Bar Vac® CD/T; Boehringer Ingelheim Vetmedica, Inc., St Joseph, MO, USA).

Animals resided at all or most times in eight nonsupplement pastures containing a shelter, with water available ad libitum in all pastures. Two animal groups were on a control treatment (CO), without supplementation other than free access to a loose mineral–vitamin mixture (16:8 Meat Maker® – 987; Sweetlix LLC, Salt Lake City, UT, USA). The guaranteed analysis was 14.0–16.8% calcium, 8% phosphorus, 10–12% salt, 1.5% magnesium, 1.5% potassium, 1.5% sulfur, 1.25% iron, 1.25% manganese, 1.25% zinc, 240 mg/kg cobalt, 1750 mg/kg copper, 450 mg/kg iodine, 50 mg/kg selenium, 661,500 IU/kg vitamin A, 110,250 IU/kg vitamin D₃, and 882 IU/kg vitamin E. Two groups received free access to a multi-nutrient supplement block (SB; Meat Maker® 20% Pressed Block, Sweetlix). The guaranteed analysis was 20% crude protein (CP), 1% crude fat, 12% crude fiber (maximum), 5–6% calcium, 3% phosphorus, 18.5–21.0% salt, 0.5% magnesium, 1.5% potassium, 0.5% sulfur, 30 mg/kg cobalt, 230 mg/kg copper, 60 mg/kg iodine, 1700 mg/kg iron, 1700 mg/kg manganese, 6.2 mg/kg selenium, 1700 mg/kg zinc, 220,500 IU/kg vitamin A, 55,125 IU/kg vitamin D₃, and 441 IU/kg vitamin E. Blocks were situated on the ground outside of the shelter, whereas the loose mineral supplement was placed in a feeder attached to the inside of shelters to minimize exposure to moisture. Two pastures included mimosa (Albizia julibrissin Durazz) trees that had been planted as seedlings in the summer of 2001. Each pasture had 10 rows of mimosa trees separated by 3.1 m and originally with a 0.46-m interval within rows. These pastures had been grazed each summer either continuously or periodically and pruned occasionally. However, pruning had not been aggressive, resulting in many leaves above the reach of goats. Animals on each side of the mimosa pastures were given periodic access to the same mimosa pastures, which were not grazed otherwise. Two groups were given 6 h of access from 09:00 to 15:00 h every Wednesday (1X) and two groups had access for 3 h/day from 09:00 to 12:00 h on each Monday and Friday (2X). During these times, animals were not allowed back into their nonsupplement pasture. Each mimosa pasture was grazed by the same 1X and 2X groups. The 1X and 2X groups had free access to the loose mineral–vitamin supplement given to CO groups.

Botanical composition (as is or fresh basis) was determined immediately before and after the experiment. A 92-m long transect was randomly placed in pastures at two sites, with readings (200 per paddock) made every 0.92 m. Plant species were categorized as bermuda grass (Cynodon dactylon), johnson grass (Sorghum halepense), other grasses, ragweed (Ambrosia artemisiifolia), and other forbs, and percentages in each paddock were calculated. After closely observing grazing, samples of forage similar to that being selectively grazed were hand-plucked in each pasture in the middle of the 4-wk periods. Samples were analyzed for dry matter (DM), ash, Kjeldahl CP (AOAC 2006), neutral detergent fiber ((NDF) with use of heat stable amylase and containing residual ash), acid detergent fiber (ADF), and acid detergent lignin (ADL; filter bag technique of ANKOM Technology Corp., Fairport, NY, USA).

Pre- and post-grazed herbage mass was assessed by clipping at a height of 2.5 cm in five randomly placed 0.25 m² quadrats. Mass of DM was determined by drying for 48 h in a forced-air oven at 55°C. The number of live mimosa trees was determined before the experiment and one month after the end. Does and kids were weighed and body condition score (BCS) of does (1–5, with 1 and 5 being very thin and fat, respectively) was determined by a panel of three individuals, as described by Ngwa et al. (2007), every 4 wk. In addition, the FAMACHA^© score (van Wyk and Bath 2002) and feces and jugular blood samples were collected. Fecal egg count (FEC) was determined with a modified McMaster technique (Whitlock 1948). After clotting blood samples were centrifuged at 1500×g for 20 min to harvest serum for analysis of immunoglobulins (Ig). Total concentrations of IgA, IgM, and IgG were determined by sandwich enzyme-linked immunosorbent assay (ELISA) according to directions of the manufacturer (Bethyl Laboratories, Inc.; Montgomery, TX, USA). Briefly, microtiter wells were coated with appropriate antibodies; coated wells were blocked with blocking solution; serum was added to wells and incubated for 60 min; horseradish peroxidase-conjugated secondary antibodies were added to the wells and incubated for 60 min; enzyme substrates were added to the wells and incubated for 30 min; and reactions were stopped by stop solution. The plates were read using a 96-well plate reader.

Daily high, low, and average ambient temperature (Ta) and relative humidity (RH), as well as rainfall, were derived at www.wunderground.com for the Guthrie, Oklahoma airport approximately 14 km from the study site. A temperature–humidity index (THI) was determined as: 0.8×Ta+(RH (Ta–14.4)/100+46.4 (Amundson et al. 2006).

Data were analyzed by SAS (1990). Forage data were analyzed with mixed model procedures (Littell et al. 1996); the random effect was pasture within treatment and the repeated measure was 4-wk period or time of measurement. Doe BW, average daily gain (ADG), and BCS and kid BW and ADG were analyzed by general linear model procedures, with a model consisting of treatment, litter size, and treatment×litter size and random effect of animal group within treatment. The interaction was not significant for any variable. Separation of treatment means was by three orthogonal contrasts for effects of supplementation (CO vs. the mean of SB, 1X, and 2X), type of supplement (SB vs. the mean of 1X and 2X), and frequency of supplement or mimosa pasture access (1X vs. 2X). The FAMACHA^© score, FEC (log+10), and Ig concentrations were analyzed with a mixed model similar to forage data, but with inclusion of litter size, treatment×litter size, and interactions among treatment, litter size, and time. No interaction involving litter size was significant, although the treatment×time interaction was significant (P<0.05) for FAMACHA^© score, FEC, and concentrations of IgA and IgM. Thus, these data were analyzed by time with the general linear models procedure, with orthogonal contrasts for means separation.

2.2. Experiment 2

Thirty-two multiparous Spanish does, each with two kids, were employed in the experiment conducted in 2011. The experiment lasted 12 wk and began 3 July when kids averaged 66 days of age (S.E.M.=0.8). Does and kids were treated for internal parasites at the beginning of the experiment with moxidectin (0.8 mg/kg BW), levamisole (14 mg/kg BW), and albendazole (19 mg/kg BW; Pfizer Animal Health, New York, NY, USA). Kids were vaccinated with Bar Vac® CD/T (Boehringer Ingelheim Vetmedica, Inc.) and Cornynebacterium pseudotuberculosis (Case-Bac; Colorado Serum Co., Denver, CO, USA). The CO and SB treatments were the same as in Experiment 1. However, the 1X treatment entailed free access to a supplement pasture for 24 h every Wednesday by opening of a gate between pastures. For the 2X treatment, animals were given free access to a supplement pasture for 6 h/day from 09:00 to 15:00 h every Monday and Friday, also via an open gate between pastures.

Many procedures were the same as in Experiment 1. However, forage mass was determined with a disk meter (Bransby et al. 1977; Brayan et al. 1989) at the beginning and end of each 4-wk period. Calibration (establishment of the relationship between compressed forage height and mass) was performed at the beginning of the experiment with 10 calibration points and clipping forage at a height of 2.54 cm. It was planned to conduct a calibration at each time of measurement; however, this was not possible because of very low precipitation and, consequently, low forage height and mass throughout the nonsupplement pastures. Forage mass calibration samples were dried in a forced-air oven for 48 h at 55°C (Aiken et al. 2006). There were 30 determinations of compressed forage height made in each pasture. Initial forage mass and that after 28 days in supplement pastures was not determined because of presence of some tall johnson grass, which resulted in a poor relationship between compressed forage height and mass. Also because of very low forage mass in periods 2 and 3, the quantity of hand-plucked forage samples collected in each pasture was very small, necessitating combining of samples by period across pastures before analyses.

Botanical composition was determined as in Experiment 1 but with 50% of the readings. The FEC was not determined. In contrast to Experiment 1, based on the FAMACHA^© score a number of animals required treatment for internal parasites during the study (i.e. score 3.5 or higher). Treatment was with the same anthelmintics used at the beginning of the experiment. Blood samples were not collected for Ig concentrations. Because of the weather conditions, resulting in very low forage mass, beginning the first day of period 2, up to 2.3 kg (air-dried) of a grass hay was offered daily in each pasture. Consumption was incomplete on most days, resulting in lower levels of feeding. Statistical analyses were conducted for Experiment 1, although litter size was not addressed.

3. Results

3.1. Experiment 1

Temperature and humidity were fairly typical of other years in central Oklahoma (Table 1). Precipitation was much greater in period 1 than in other periods, although small-to-moderate amounts were recorded in each period.

Table 1. Weather conditions during four 4-wk periods of grazing by Boer does and kids with different supplement treatments (Experiment 1).

	Period 1		Period 2		Period 3		Period 4
Item	Mean	S.E.M.	Mean	S.E.M.	Mean	S.E.M.	Mean	S.E.M.
Temperature (°C)
High	31.6	0.53	35.3	0.62	32.9	0.97	29.8	0.80
Low	20.4	0.59	22.9	0.40	21.9	0.47	17.5	0.83
Average	25.9	0.49	29.0	0.45	27.2	0.67	23.4	0.71
Humidity (%)
High	85.5	2.09	75.7	2.26	85.0	1.82	87.6	0.80
Low	47.4	1.69	37.1	2.24	45.8	3.05	44.6	3.08
Average	67.4	1.89	56.8	2.25	67.6	2.43	68.9	1.62
THI
High	86.2	0.78	90.1	0.71	88.2	1.37	83.7	1.30
Low	65.5	0.74	67.8	0.43	67.2	0.51	62.0	1.02
Average	74.6	0.57	77.6	0.37	76.4	0.72	71.3	1.05
Precipitation (cm)	15.7		4.3		6.2		5.2

THI=temperature-humidity index.

There was not an effect of treatment or an interaction between treatment and time of measurement on mass or composition of forage in nonsupplement pastures (P>0.05; Table 2). Mass was less at the end of periods 3 and 4 compared with other times. However, values at the last two times were moderate to high and, based on research reviewed by AFRC (1998), probably not restrictive to feed intake. Forage mass in supplement pastures did not differ among periods, and values were greater than in nonsupplement pastures except at the beginning of the experiment.

Table 2. Forage conditions during four 4-wk periods of grazing by Boer does and kids with different supplement treatments (Experiment 1).

	Period
Item	Initial	1	2	3	4	S.E.M.
Nonsupplement pasture forage mass (kg/ha)	3477^Footnoteb	3448^Footnoteb	3353^Footnoteb	2802^Footnotea	2423^Footnotea	167.7
Supplement pasture forage mass (kg/ha)	3227	4616	4656	4563	5316	682.1
Forage composition (% dry matter)
Ash		8.6^Footnotea	10.2^Footnoteb	9.6^Footnoteb	9.8^Footnoteb	0.69
Crude protein		13.0^Footnotea	14.7^Footnotea	15.1^Footnotea	19.5^Footnoteb	0.84
Neutral detergent fiber		69.2	65.4	65.6	67.4	3.17
Acid detergent fiber		35.3	35.3	36.1	34.6	1.15
Acid detergent lignin		7.7^Footnotea	9.7^FootnotebFootnotec	9.3^Footnoteb	10.4^Footnotec	0.48
Nonsupplement pasture botanical composition (%, fresh basis)
Bermuda grass	15.4^Footnotea				61.5^Footnoteb	5.84
Johnson grass	1.1				0.0	0.32
Other grasses	35.0^Footnoteb				7.9^Footnotea	3.47
Ragweed	12.6				23.4	6.38
Other forbs	35.9^Footnoteb				7.2^Footnotea	1.84
Supplement pasture botanical composition (%, fresh basis)
Bermuda grass	12.3				45.3	12.43
Johnson grass	16.3				23.3	16.94
Other grasses	22.3				4.0	2.13
Ragweed	10.8				23.3	3.91
Other forbs	38.5				4.3	14.90

^a,b,c Means in a row without a common letter differ (P<0.05).

Concentrations of ash and CP differed (P<0.05) among periods (Table 2). The ash concentration was lowest among periods (P<0.05) in period 1 and the concentration of CP was greatest (P<0.05) in period 4. The NDF and ADF concentrations were similar among periods, although the level of ADL was lowest and highest among periods (P<0.05) in periods 1 and 4, respectively.

The average number of live mimosa trees in supplement pastures was 861 and 843 at the beginning and end of the experiment. Mimosa tree leaves remained in reach of does and kids throughout the experiment. Period did not affect (P>0.05) mimosa leaf composition; mimosa leaves averaged 6.3% ash (S.E.M.=0.14), 23.4% CP (S.E.M.=1.54), 33.1% NDF (S.E.M.=0.73), 23.4% ADF (S.E.M.=1.07), and 10.8% ADL (S.E.M.=0.70).

Intake of the supplement block averaged 111 g/day (air-dried) per doe, which was slightly less than observed previously by Goetsch et al. (2007) with Spanish does each nursing two kids in the same pasture area (i.e. 116 g/day; DM). The range in intake listed by the manufacturer is 57–114 g/day per animal (air-dried). There was no effect of treatment on level of intake of loose mineral supplement, with means of 8.7, 9.9, and 8.7 g/day (air-dried) per doe for CO, 1X, and 2X, respectively (S.E.M.=0.43). Values were also slightly less than noted by Goetsch et al. (2007; i.e. 12.3 g/day per doe, DM).

As expected based on times of the year when measured, botanical composition in nonsupplement pastures differed markedly between the beginning and end of the experiment. Bermuda grass (Cynodon dactylon) was the most prevalent plant species at the end of the experiment, but other grasses and forbs were more abundant at the beginning. A nonsignificant increase in prevalence of ragweed (Ambrosia spp.) occurred from the beginning to end. The major difference in botanical composition between nonsupplement and supplement pastures was in a greater level of johnson grass (Sorghum halepense) in the latter.

Neither doe BW nor ADG was affected by treatment (P>0.05; Table 3). Doe BW was less (P<0.05) for litter size 2 than 1 throughout the experiment. However, the magnitude of difference was less at the end of the experiment than the beginning, in accordance with greater BW loss for litter size 1 than 2 in periods 3 and 1–4 (P<0.05). Doe BCS and its change during the experiment were generally in accordance with differences between litter sizes in BW and ADG.

Table 3. Performance of Boer does and kids grazing grass/forb pastures with different supplement treatments (Experiment 1).

		Treatment^Footnote1					Contrast P value^Footnote2			Litter size
Item^Footnote3	Period	CO	SB	1X	2X	S.E.M.	Supp	Type	Freq	1	2	S.E.M.
Doe BW	Initial	43.1	42.0	43.5	42.1	1.21	0.718	0.627	0.449	46.0^Footnoteb	39.3^Footnotea	1.10
	1	45.2	43.6	43.9	43.5	0.98	0.254	0.941	0.767	47.1^Footnoteb	41.0^Footnotea	1.06
	2	44.6	42.8	43.8	43.4	1.46	0.481	0.663	0.843	46.7^Footnoteb	40.6^Footnotea	0.98
	3	41.5	38.3	41.2	41.3	1.09	0.371	0.091	0.970	43.2^Footnoteb	37.9^Footnotea	0.91
	4	40.4	37.3	40.9	40.4	0.99	0.485	0.049	0.738	42.2^Footnoteb	37.3^Footnotea	0.85
Doe ADG	1	76	58	16	51	3.3	0.426	0.581	0.492
	2	−20	−31	−4	−4	3.1	0.861	0.521	1.000
	3	−112	−161	−94	−76	15.8	0.939	0.017	0.469	−126^Footnotea	−95^Footnoteb	9.3
	4	−38	−33	−9	−3	19.6	0.565	0.577	0.508
	1–4	−23	−42	−23	−15	11.5	0.830	0.174	0.633	−33^Footnotea	−19^Footnoteb	3.9
Doe BCS (1–5)	Initial	2.46	2.51	2.71	2.52	0.065	0.182	0.262	0.112	2.85^Footnoteb	2.24^Footnotea	0.072
	3	2.41	2.14	2.32	2.43	0.056	0.161	0.025	0.259	2.43^Footnoteb	2.22^Footnotea	0.037
	4	2.48	2.16	2.53	2.61	0.083	0.663	0.015	0.518	2.54^Footnoteb	2.35^Footnotea	0.046
	Change	0.02	−0.35	−0.18	0.09	0.100	0.221	0.063	0.127	−0.31^Footnotea	0.10^Footnoteb	0.072
Kid BW	Initial	6.0	7.8	7.5	6.5	0.37	0.046	0.174	0.148
	1	10.6	12.9	12.3	11.3	0.52	0.063	0.159	0.261	12.8^Footnoteb	10.7^Footnotea	0.42
	2	14.1	16.0	15.6	15.3	0.58	0.078	0.533	0.749	17.0^Footnoteb	13.5^Footnotea	0.43
	3	16.2	17.1	17.6	17.8	0.56	0.110	0.414	0.768	19.3^Footnoteb	15.0^Footnotea	0.46
Kid ADG	1	163	183	171	170	6.6	0.202	0.185	0.910	199^Footnoteb	146^Footnotea	4.7
	2	124	110	121	145	5.4	0.822	0.027	0.032	150^Footnoteb	100^Footnotea	4.8
	3	75	39	69	88	10.8	0.496	0.040	0.281	84^Footnoteb	52^Footnotea	3.9
	1–3	121	111	120	134	3.3	0.759	0.015	0.038	144^Footnoteb	99^Footnotea	3.0

¹ CO=control; SB=supplement block; 1X=access on 1 day for 6 h to supplement pasture with mimosa trees; 2X=access for 3 h/day on 2 days to supplement pasture.

² Supp=supplementation (CO vs. mean of SB, 1X, and 2X; Type=supplement type (SB vs. mean of 1X and 2X); Freq=frequency of supplement pasture access (1X vs. 2X).

³ BW=body weight; ADG=average daily gain; BCS=body condition score (1 and 5=extremely thin and fat, respectively).

^a,b Litter size means without a common letter differ (P<0.05).

There were effects of supplement type on kid ADG in periods 2, 3, and 1–3 and of frequency of access to supplement pastures in periods 1 and 1–3 (P<0.05; Table 3). These effects were primarily due to relatively high kid ADG for the 2X treatment. Kid BW and ADG were greater for litter size 1 than 2 throughout the experiment, with the magnitude of difference in BW increasing as the experiment progressed. The difference in ADG was smaller in period 3 than in periods 1 and 2 presumably because of greater forage intake late than early in the experiment.

There were treatment×time or litter size×time interactions (P<0.05) in FAMACHA^© score, FEC, and concentrations of IgA and IgM, but not for level of IgG. There were no treatment effects on FAMACHA^© score or FEC at the different times (P>0.05; Table 4). At the end of the experiment, FAMACHA^© score was less for litter size 1 than 2 (P<0.05). The FEC was greater for does with litter size 1 than 2 at the beginning of the experiment and end of period 1 (P<0.05). There were effects of supplementation on the IgA concentration at the end of periods 3 (P<0.05) and 4 (P<0.10), and the same was true for type of supplementation (P<0.05). These effects were due to lower values for 1X and 2X compared with CO and SB. There were a number of treatment effects on the concentration of IgM. At the end of period 1, the concentration of IgM was lower with supplementation than without supplementation (P<0.05), for 1X and 2X vs. SB, and for 2X than for 1X (P<0.07). Likewise, the IgM concentration was lower with supplementation than without supplementation at the end of periods 3 and 4 (P<0.05), and after period 3 the level was lower for 2X vs. 1X (P<0.05). Does with 1 kid had a lower IgM concentration after periods 1, 3, and 4 than did does with 2 kids (P<0.05). Treatment did not affect the concentration of IgG, but the level ranked (P<0.05) period 2<3<4<1 (30.1, 16.0, 24.8, and 27.2 mg/ml for CO, SB, 1X, and 2X, respectively; S.E.M.=0.77).

Table 4. FAMACHA^© score, FEC, and Ig concentrations for Boer grazing grass/forb pastures with different supplement treatments (Experiment 1).

		Treatment^Footnote1					Contrast P value^Footnote2			Litter size
Item^Footnote3	Period	CO	SB	1X	2X	S.E.M.	Supp	Type	Freq	1	2	S.E.M.
FAMACHA^© score (1–5)	0	2.25	2.13	2.25	2.25	0.337	0.920	0.777	1.000
	1	3.13	3.38	3.13	3.38	0.280	0.633	0.734	0.561
	2	3.25	3.38	3.50	3.38	0.293	0.648	0.870	0.778
	3	3.38	3.88	3.38	3.75	0.300	0.447	0.443	0.426
	4	3.00	3.25	2.88	3.13	0.405	0.867	0.641	0.685	2.88^Footnotea	3.26^Footnoteb	0.122
FEC (log10 number/g+10)	0	5.54	5.43	5.41	5.54	0.267	0.823	0.879	0.749	6.07^Footnoteb	4.89^Footnotea	0.249
	1	7.85	8.31	7.77	8.02	0.351	0.674	0.391	0.646	8.19^Footnoteb	7.78^Footnotea	0.124
	2	7.45	8.51	7.76	7.52	0.289	0.213	0.068	0.608
	3	7.76	8.21	8.02	7.85	0.228	0.373	0.379	0.639
	4	6.18	6.39	6.51	5.96	0.461	0.853	0.796	0.444
FEC (number/g)	0	243	217	215	245					421	123
	1	2551	4046	2361	3025					3602	2382
	2	1701	4969	2326	1842
	3	2344	3671	3022	2563
	4	473	586	663	376
IgA (mg/ml)	0	57	41	49	53	3.3	0.068	0.086	0.428	40^Footnotea	60^Footnoteb	5.1
	1	68	66	69	51	5.1	0.328	0.389	0.068
	2	91	89	74	96	16.8	0.858	0.965	0.395
	3	165	154	112	118	10.7	0.041	0.048	0.690
	4	159	183	113	115	9.0	0.010	0.004	0.869
IgM (mg/ml)	0	1441	1292	1503	1326	174.6	0.999	0.743	0.504
	1	1256	1153	988	813	49.4	0.008	0.016	0.067	950^Footnotea	1155^Footnoteb	49.3
	2	1295	1109	965	1111	112.5	0.139	0.646	0.414
	3	1333	958	979	485	115.0	0.015	0.196	0.039	844^Footnotea	1033^Footnoteb	58.5
	4	1720	1125	1001	816	176.8	0.021	0.390	0.500	977^Footnotea	1354^Footnoteb	66.8
IgG (mg/ml)	Mean	25.5	24.6	23.0	25.0	0.81

¹ CO=control; SB=supplement block; 1X=access on 1 day for 6 h to supplement pasture with mimosa trees; 2X=access for 3 h/day on 2 days to supplement pasture.

² Supp=supplementation (CO vs. mean of SB, 1X, and 2X; Type=supplement type (SB vs. mean of 1X and 2X); Freq=frequency of supplement pasture access (1X vs. 2X).

³ FEC=fecal egg count; Ig=immunoglobin.

^a,b Litter size means without a common letter differ (P<0.05).

3.2. Experiment 2

Times of the year for periods 1, 2, and 3 correspond to periods 2, 3, and 4 in Experiment 1, respectively. In this regard, temperature was greater in Experiment 2 than 1, although humidity was less, resulting in THI slightly greater in Experiment 2 (Table 5). Precipitation was also less in Experiment 2, and that before the experiment was also minimal. As a consequence, forage mass in nonsupplement pastures was much lower in Experiment 2 (Table 6) vs. 1. Forage mass was not influenced by treatment and there was not a treatment×time interaction. Mass ranked (P<0.05) initial>period 1>periods 2 and 3, with the greatest magnitude of difference between the first two times (i.e. 688 kg/ha). The level of forage mass in supplement pastures was slightly greater than in nonsupplement pastures at the end of periods 3 and 4. Forage composition was not markedly different from that in Experiment 1, other than a lower level of CP in the last period compared with Experiment 1. Grass hay samples averaged 7.2% ash (S.E.M.=0.46), 6.8% CP (S.E.M.=1.47), 67.4% NDF (S.E.M.=2.43), 43.2% ADF (S.E.M.=1.81), and 8.0% ADL (S.E.M.=0.22).

Table 5. Weather conditions during three 4-wk periods of grazing by Spanish does and kids with different supplement treatments (Experiment 2).

	Period 1		Period 2		Period 3
Item	Mean	S.E.M.	Mean	S.E.M.	Mean	S.E.M.
Temperature (°C)
High	40.9	0.35	39.1	0.66	30.7	1.22
Low	27.1	0.32	25.0	0.59	16.4	1.17
Average	33.7	0.28	31.9	0.58	23.3	1.12
Humidity (%)
High	63.6	2.08	69.6	3.14	73.9	3.37
Low	22.9	0.78	27.2	1.74	27.9	2.22
Average	41.1	1.43	47.9	2.42	49.1	2.82
THI
High	95.8	0.54	94.5	0.77	82.3	1.56
Low	70.8	0.29	69.1	0.54	59.7	1.18
Average	81.2	0.20	80.0	0.47	68.8	1.21
Precipitation (cm)	0.5		4.5		2.7

THI=temperature-humidity index.

Table 6. Forage conditions during three 4-wk periods of grazing by Spanish does and kids with different supplement treatments (Experiment 2).

	Period
Item	Initial	1	2	3	S.E.M.
Nonsupplement pasture forage mass (kg/ha)	1530^Footnotec	842^Footnoteb	791^Footnotea	750^Footnotea	42.3
Nonsupplement pasture forage height (cm)	5.0^Footnotec	2.2^Footnoteb	1.9^Footnotea	1.8^Footnotea	0.18
Supplement pasture forage mass (kg/ha)			1185	955	92.3
Forage composition (% dry matter)
Ash		8.2	8.7	7.6
Crude protein		12.7	16.2	13.4
Neutral detergent fiber		68.3	59.7	64.2
Acid detergent fiber		35.6	33.6	35.3
Acid detergent lignin		8.0	7.3	7.3
Mimosa composition (% dry matter)
Ash		7.7	7.0	7.6	0.45
Crude protein		21.4^Footnotea	25.9^Footnoteb	26.6^Footnoteb	1.46
Neutral detergent fiber		42.1^Footnoteb	30.2^Footnotea	29.8^Footnotea	1.63
Acid detergent fiber		25.7	22.5	19.7	1.11
Acid detergent lignin		11.4	9.0	7.9	0.89
Nonsupplement pasture botanical composition (%, fresh basis)
Bermuda grass	69.0^Footnotea			89.2^Footnoteb	2.59
Johnson grass	1.6			0.0	0.57
Other grasses	2.4			0.5	0.95
Ragweed	12.5^Footnoteb			2.9^Footnotea	2.50
Other forbs	14.5^Footnoteb			7.4^Footnotea	2.50
Supplement pasture botanical composition (%, fresh basis)
Bermuda grass	45.0			62.0	18.07
Johnson grass	24.5			18.5	17.61
Other grasses	2.5			0.5	1.80
Ragweed	12.0			6.0	6.96
Other forbs	16.0			13.0	8.06

^a,b,c Means in a row without a common superscript letter differ (P<0.05).

The average number of live mimosa trees counted was 752 and 671 at the beginning and end of the experiment, respectively, with leaves available for intake by does and kids throughout the experiment. There was some stripping of tree bark by goats. But because the second determination was after the study rather than in the subsequent spring, it is not known if bark stripping caused tree death. Probably it did not, because these trees regrow if above ground growth is clipped or severely damaged until root reserves are exhausted. Concentrations of CP and NDF in mimosa leaves varied (P<0.05) periods, with the lowest level of CP and highest NDF in period 1 (P<0.05; Table 6). Levels of ADF and ADL were similar among periods.

Intake of the supplement block averaged 104.2 g/day per doe (air-dried). Botanical composition of pastures at the beginning of the experiment (Table 6) was considerably different from that in Experiment 1, which relates in part to the later start date and possibly differences in temperature and precipitation. These conditions presumably also contributed to other differences in the botanical composition at the end of the experiment, such as a higher level of bermuda grass in both nonsupplement and supplement pastures.

There were no treatment effects on doe BW (P>0.05; Table 7). Somewhat similar to Experiment 1, ADG was positive in period 1 but thereafter BW generally declined. The ADG was greater with supplementation than without supplementation in periods 3 and 1–3 (P<0.05). Also, ADG was greater for 1X and 2X than for SB in periods 3 (P<0.05) and 1–3 (P<0.09). Likewise, change in doe BCS was greater for 1X and 2X vs. SB (P<0.06). There were no treatment effects on the FAMACHA^© score and anthelmintic treatment of does and kid BW, ADG, FAMACHA^© score, and anthelmintic treatment.

Table 7. Performance, FAMACHA^© score, and anthelmintic treatment of Spanish does and kids grazing grass/forb pastures with different supplement treatments (Experiment 2).

		Treatment^Footnote1					Contrast P value^Footnote2
Item^Footnote3	Period	CO	SB	1X	2X	S.E.M.	Supp	Type	Freq
Doe BW (kg)	Initial	41.8	41.3	41.4	41.3	1.39	0.807	0.986	0.928
	1	45.2	44.1	43.9	43.4	0.95	0.272	0.689	0.728
	2	40.5	40.5	39.9	40.3	1.47	0.890	0.845	0.856
	3	38.1	38.6	39.5	40.3	1.29	0.409	0.454	0.703
Doe ADG (g)	1	123	100	87	76	24.2	0.279	0.557	0.761
	2	−167	−129	−141	−109	31.9	0.329	0.915	0.527
	3	−87	−69	−16	−2	14.3	0.025	0.026	0.544
	1–3	−44	−33	−23	−12	5.5	0.028	0.086	0.224
Doe BCS (1–5)	Initial	2.44	2.47	2.47	2.41	0.102	0.934	0.816	0.689
	3	2.55	2.52	2.67	2.84	0.041	0.050	0.008	0.040
	Change	0.11	0.05	0.20	0.44	0.082	0.276	0.053	0.114
Doe FAMACHA^© score (1–5)	Initial	3.75	3.63	3.88	3.63	0.198	0.864	0.633	0.422
	1	3.75	3.81	3.81	3.81	0.256	0.843	1.000	1.000
	2	3.56	3.44	3.63	3.44	0.248	0.838	0.773	0.621
	3	3.25	3.13	3.25	3.19	0.031	0.158	0.071	0.230
Doe anthelmintic treatment (%)	1	75.0	87.5	87.5	75.0	19.76	0.734	0.809	0.678
	2	62.5	62.5	62.5	62.5	0.217	1.000	1.000	1.000
	3	25.0	37.5	50.0	50.0	6.25	0.158	1.000	0.047
Kid BW (kg)	Initial	15.0	13.4	14.0	14.3	0.89	0.364	0.521	0.798
	1	18.4	16.2	16.8	17.5	1.17	0.313	0.523	0.698
	2	18.1	16.5	17.0	17.3	1.09	0.410	0.649	0.820
Kid ADG (g)	1	123	99	102	114	11.8	0.262	0.595	0.504
	2	−11	11	5	−7	11.5	0.349	0.434	0.532
	1–2	56	55	53	54	6.5	0.817	0.819	0.955
Kid FAMACHA^© score (1–5)	1	3.19	3.38	3.34	3.22	0.141	0.487	0.617	0.566
	2	2.88	3.09	3.31	3.19	0.090	0.036	0.228	0.381
Kid anthelmintic treatment (%)	1	37.5	50.0	37.5	37.5	0.242	0.889	0.695	1.000
	2	12.5	18.8	25.0	37.5	11.27	0.325	0.416	0.477

¹ CO=control; SB=supplement block; 1X=access on 1 day for 6 h to supplement pasture with mimosa trees; 2X=access for 3 h/day on 2 days to supplement pasture.

² Supp=supplementation (CO vs. mean of SB, 1X, and 2X; Type=supplement type (SB vs. mean of 1X and 2X); Freq=frequency of supplement pasture access (1X vs. 2X).

³ BW=body weight; ADG=average daily gain; BCS=body condition score (1 and 5=extremely thin and fat, respectively).

4. Discussion

4.1. Supplement pasture access

Periodic access to supplement pastures with mimosa trees had relatively small influence on animal performance, with only slight increases in doe ADG in Experiments 1 and 2, respectively. For Experiment 1, little effect may have resulted from moderate-to-high forage mass and nutritive value throughout the experiment. This was not the case in Experiment 2 in regards to forage mass. It is possible that effects of 1X and 2X, and possibly SB, would have been greater in Experiment 2 if grass hay had not been supplemented. Hay was given from an animal care concern, as it was projected that morbidity and/or mortality could have occurred without doing so. Thus, conditions of these experiments were not designed to simulate severe drought or dry season conditions more common in other areas of the world where pasture or range vegetation is relied upon regardless of availability.

Results of Experiment 1 suggested that the effects of 1X and 2X treatments might have been greater with longer periods of access to supplement pastures, which led to changes made for Experiment 2. The lack of marked effect on animal performance of 1X and 2X treatments also observed in Experiment 2 probably relates to the ability of goats to select vegetation in nonsupplement pastures considerably higher in nutritive value that the average of vegetation available (Baumont et al. 2000), as well as grass hay that was supplemented. Nonetheless, although the magnitude of effect of supplement pasture access in periods 1–3 was not substantial, it was quite large in period 3 after kids were weaned. This could imply that in some manner the presence of nursing kids prevented effects of 1X and 2X treatments. For example, perhaps nursing of kids limited time for consumption by does of mimosa tree leaves, the likelihood of which could have been increased by high temperature that, from casual observation, minimized grazing from mid-morning to late afternoon. Additional measures would have been, or further research will be, required to characterize impact of kid presence.

Overall, findings of both experiments indicate that the attributes of protein banks in areas with severe drought or dry seasons and little or no other available feedstuffs high in nutritive value (Paladines 1984) may not be realized in other settings in which there is infrequent access to protein banks and nutritional conditions are more favorable, such as in these two experiments. Access to supplement pastures more frequent than once or twice weekly, such as daily, was a consideration. For example, Moron-Fuenmayor and Clavero (1999) gave West African lambs grazing buffel grass (Cenchrus ciliaris) pastures daily access to pasture with leucaena (Leucaena leucocephala) for 2 h and increased BW by 22% compared with grazing buffel grass only. However, the SB treatment in the present experiments provided daily consumption of supplemental nutrients. In previous experiments with sequential access of kids and lambs to 25% of these supplement pastures with mimosa every 1–2 wk, all available mimosa leaves were consumed within the first few days (Animut et al. 2007; Goetsch et al. 2007; Yiakoulaki et al. 2007). Thus, it was felt that an unreasonably low stocking rate or very short daily periods of access would be necessary to maintain availability of mimosa tree leaves throughout the experiments. Furthermore, the labor required and its cost in some regions and countries may deem use of such treatments unlikely. But perhaps access on alternate days or three times weekly (i.e. Monday, Wednesday, and Friday), as often employed for protein supplementation of wintering beef cows (McCollum and Horn 1990), is worthy of attention as a compromise between minimal labor input and achieving an adequate quantity and frequency of consumption of supplemental nutrients.

4.2. Supplement block

In neither experiment was there benefit from use of the supplement block, in agreement with a previous experiment conducted in the same pasture area (Goetsch et al. 2007). This suggests, as alluded to above, that the nutritional plane with the CO treatment did not severely restrict performance. Also, based on levels of supplement block intake, though in the upper range listed by the manufacturer, and composition values of Goetsch et al. (2007), levels of supplemental CP and digestible energy consumed by does were no more than 10% of requirements.

4.3. Litter size

Birth weight is typically greater with litter size 1 than 2 or more (AFRC 1998). However, initial kid BW in Experiment 1 at 22 days of age was similar between litter sizes. Greater BW loss for does with 1 vs. 2 kids was despite total litter BW gain 54 g/day less for litter size 1. Most likely, this was a function of greater initial doe BW for litter size 1 than 2, concomitant with greater tissue available for mobilization. Hence, less doe BW loss for litter size 2 probably resulted from both greater tissue mobilization after birth before the experiment began and in late gestation. The ADG of kids in Experiment 2 was approximately 20 g/day less than that of kids with litter size 2 in periods 2 and 3 of Experiment 1, which likely was a consequence of differences in forage mass, temperature, and perhaps level of internal parasitism in does and kids and/or Boer and Spanish kids in Experiment 1 and 2, respectively.

4.4. Internal parasitism

At least two factors may have been involved in considerable treatment of animals with anthelmintics in Experiment 2 in contrast to Experiment 1. One is that relatively low forage mass and height in Experiment 2 caused high ingestion of infective Haemonchus contortus larvae relative to Experiment 1 in which forage mass and height were greater; however, lower precipitation and higher temperature in Experiment 2 should have lessened effects of forage conditions (Hoste and Torres-Acosta 2011). Secondly, anthelmintic resistance was probably greater in Experiment 2, conducted in 2011, than in 2008 when Experiment 1 occurred. In support, the combination of three anthelmintics used in Experiment 2 did not appear particularly efficacious. Potential differences between Boer and Spanish breeds in susceptibility to internal parasites cannot be evaluated since the experiments were in different years.

Protein intake can affect internal parasitic infections (Sykes and Coop 2001). For example, Bawden (1969) found that sheep fed an alfalfa-based diet, high in protein, harbored less nematodes at 56 days after infection compared with ones fed a straw-based diet low in protein. Similarly, Singh et al. (1995) showed that the worm burden in goats was lower with than without cottonseed meal supplementation. The lack of effect of treatment in Experiment 1 may reflect smaller differences in protein intake than in these previous studies.

Although 1X and 2X treatments did not have major effect on animal performance in Experiment 1, it appears that the immunity was impacted. Higher concentrations of IgA and IgM during periods 3 and 4 for 1X and 2X compared with the control implies suboptimal protein intake by control animals. Protein malnutrition in animals impairs the transportation of Ig from blood to target tissues, thereby resulting in elevated blood levels (McGee and McMurray 1988; Sullivan et al. 1993). But, protein deficiency alters the populations of immune cells and reduces Ig production (Lopez et al. 1985; Manhart et al. 2000; Vidueiros et al. 2008). The results of Experiment 1 may indicate that marginal intake of protein can be detrimental to the immune system without affecting BW.

5. Conclusion

Use of the multi-nutrient block under these conditions was not beneficial. There were only relatively small improvements of infrequent access to supplement pastures with the leguminous mimosa tree. Forage availability and nutritive value did not appear sufficiently limiting to facilitate substantial improvements in performance of meat goats from the different supplement treatments. This may also relate to a high degree of vegetation selection of goats with moderate and even low levels of forage mass. Future research should be conducted with longer and/or more frequent periods of access to supplement pastures with leguminous trees.