Fermentative Profile, Chemical Composition and In Situ Rumen Degradability of Capiaçu Elephant Grass Silage Wilted or with Added Cornmeal

Abstract

Silage is an essential tool for maintaining productivity, especially during the dry season and when pasture availability is limited. However, it is necessary to establish increasingly efficient methods for producing this feed, seeking to minimize losses and provide maximum nutritional benefit. This study aimed to evaluate the quality of Cenchrus purpureus (Schumach.) Morrone cv. BRS Capiaçu silage with cornmeal addition or after 3 or 5 days of wilting, focusing on fermentative profile, chemical composition, and in situ degradation. A completely randomized design with four treatments and three replicates was used: Control (CON), ensilage with 8% ground cornmeal (SGC), wilting for 3 days (WI3), and wilting for 5 days (WI5), totaling 12 silos. After 120 days, the silages were analyzed for pH, volatile fatty acids, chemical composition, and rumen degradability using three cannulated cows. Data were subjected to ANOVA and Tukey’s test (p < 0.05). The SGC and WI3 treatments showed lower pH (4.55 and 4.52) and butyric acid (0.27 and 0.33%) and higher lactic acid (2.32 and 1.57%) contents compared with CON and WI5 (p < 0.001). They also presented higher dry matter (257.2 and 318.3 g/kg) and crude protein (63.8 and 58.5 g/kg) and lower fiber fractions (p < 0.001). For rumen degradability, SGC had the highest values for fraction “A” and effective degradability of dry matter at 5 and 8%/h passage rates (p = 0.001). Cornmeal addition and 3-day wilting improved silage quality, but only cornmeal enhanced degradability.

1. Introduction

The Brazilian livestock industry uses pastures as the main source of feed for ruminants [1]. However, tropical regions present seasonal variations, reducing the availability and quality of pastures, and impacting the profitability of production systems [2]. In this scenario, silage is a fundamental tool to maintain good production rates, particularly during periods of feed scarcity [3,4,5].

Elephant grass cv. BRS Capiaçu (Cenchrus purpureus (Schumach.) Morrone cv. BRS Capiaçu) is a forage with high potential for biomass production, capable of producing up to 50 ton/ha/year of dry matter (DM) [6]. However, ensiling process of tropical grasses is associated with challenges, such as low DM concentration (<300 g/kg DM), low soluble carbohydrate content (<40 g/kg DM), and high buffering capacity (>40 g of lactic acid/kg DM), which affects the quality of the silage [7,8,9].

To improve the quality of silage, additives such as bacterial inoculants or moisture sequestrant may be used [10,11]. Ground cornmeal, as a moisture sequestrant, reduces gas and effluent losses and increases DM recovery [12,13,14]. However, the addition of ground cornmeal can increase the cost of silage production.

Alternatively, wilting can achieve an adequate DM content [15,16], improving fermentation quality [17], and reducing losses due to gases and effluents [16]. Nevertheless, the quality of the silage can be impaired if the harvested forage is exposed to the sun for a prolonged period [18,19,20].

In this context, there is a lack of studies on the ensiling of elephant grass cv. BRS Capiaçu, particularly regarding ensiling methods. These include the use of moisture sequestrant and wilting over different periods, including 3 and 5 days. Therefore, it is necessary to develop efficient methods for the conservation of this forage to improve the quality of fermentation and thus the nutritional value of the silage produced.

This study was prepared under the hypotheses that: (1) the addition of ground cornmeal during the ensiling process promotes adequate fermentation of BRS Capiaçu elephant grass silage, resulting in a feed with better nutritional value, (2) the use of wilting grass for 3 or 5 days improves silage quality. Thus, the objective was to evaluate the effects of adding ground cornmeal as a moisture sequestrant or wilting grass for 3 or 5 days on the fermentation characteristics, chemical composition, and in situ degradation of BRS Capiaçu elephant grass silage.

2. Materials and Methods

All procedures involving animals were previously reviewed and approved by the Ethics Committee for the Use of Production Animals of the Federal University of Viçosa, under protocol number 44/2024. During the experiment, the animals were maintained on a specific diet with ad libitum access to water and food, were housed in a compost barn system equipped with fans and maintained full productive life during and after the experiment.

2.1. Study Location, Design, and Treatments

The study was carried out at the Teaching, Research, and Extension Dairy Farm of the Federal University of Viçosa (UFV), Viçosa, Minas Gerais, Brazil (20°45′14″ S, 42°52′55″ W). The region has a climate classified as Cwa (mesothermal) according to the [21], characterized by hot and rainy seasons (spring–summer) from October to March and cold and dry seasons (autumn–winter) from April to September, with an average temperature of 20.4 °C and an average annual rainfall of 1251 mm [22]. During the days in which the ensiling processes were carried out, the following data on total precipitation, average temperature, and average relative humidity were recorded (Table 1). These data were obtained by the automatic meteorological station in Viçosa, located 2.8 km away from the experiment [23].

Table 1. Data on maximum and minimum temperatures, average relative humidity, and daily rainfall from the weather station in Viçosa during the harvest and wilting periods.

Days	Activity Performed	Max. Temperature (°C)	Min. Temperature (°C)	Air Humidity (%)	Precipitation (mm)
1	Harvesting of all treatments; Beginning of wilting of WI3 and WI5; Grinding and ensiling of grass from CON and SGC treatments	22.9	8.8	81.0	0
2	Wilting of grass in treatments WI3 and WI5	23.6	8.7	80.6	0.2
3	Wilting of grass in treatments WI3 and WI5	24.4	11.9	81.1	0
4	Grinding and ensiling of grass from WI3 treatment and wilting from WI5	24.2	12.9	79.0	0
5	Wilting of grass in treatments WI5	23.6	12.4	77.8	0
6	Grinding and ensiling of grass from the WI5 treatment	22.7	10	84.9	0.2

The study was conducted in a completely randomized design evaluating four treatments with three replicates each: (i) control (CON; grass ensiled immediately after harvesting); (ii) grass ensiled after cutting with the addition of ground cornmeal (SGC; 8% in relation to the mass of natural matter); (iii) grass ensiled after wilting for three days (WI3); iv) grass ensiled after wilting for five days (WI5).

2.2. Sample Collection, Processing, and Ensiling

The elephant grass cv. BRS Capiaçu was harvested in winter at about 150 days of age and an average height of 4 m, which is within the recommended range to ensure good nutritional quality and productivity [6], and it presented approximately 22% DM. The grass was planted in an area of 0.7 ha, with a between rows of 1.30 m, and the number of seedlings used for the study was 6 tons/ha. The ensiling of treatments WI3 and WI5 was carried out at predetermined times to ensure that both treatments reached the established wilting time.

All treatments were harvested manually with a sickle, with the cut being made at a height of about 15 cm above ground. The grass was chopped to 20 mm in a silage chopper (Guarnieri MG—350), and then the Magniva Basic inoculant (Lallemand) was added according to the manufacturer’s recommendations, at a rate of 100 g per 50 tons of forage. This inoculant consists of a combination of the Pediococcus acidilactici CNCM I-4622, with minimum-guarantee levels of 2.5 × 10¹⁰ UFC/g and Lactobacillus plantarum CNCM I-3736, with minimum-guarantee levels of 2.5 × 10¹⁰ UFC*/g strains. To evaluate the quality of the machine cut and mixtures with cornmeal, the Penn State method was used to check the size of the particles produced (Table 2) [24].

Table 2. Particle distribution in Penn State evaluation methodology of Cenchrus purpureus (Schumach.) Morrone cv. BRS Capiaçu grass with the addition of cornmeal and different wilting periods.

Sieves	Treatments 1
	CON	SGC	WI3	WI5
19 mm (%)	9.45	5.04	11.81	8.27
8 mm (%)	56.24	57.92	54.20	48.10
4 mm (%)	29.50	28.78	27.43	32.97
Pan (%)	4.81	8.26	6.56	10.66

¹ CON = control; SGC = additional 8% of cornmeal; WI3 = wilting for 3 days; WI5 = wilting for 5 days.

In the SGC treatment, cornmeal was added at 8% by mass of the fresh matter and manually mixed, aiming to increase the DM content of the material to 25%, bringing it within the recommended range of 25 to 35% for ensiling [25]. The wilting of elephant grass cv. BRS Capiaçu was done by spreading the plant in thin layers, allowing air to circulate and promoting moisture loss. The exposure time of the grass under these conditions varied according to treatments WI3 (3 days) and WI5 (5 days), as described in Table 1.

After obtaining the material from each treatment, the material was ensiled in 10-litre experimental silos made of sealed plastic buckets. These experimental silos had a height of 29.5 cm, a top diameter of 24 cm and a bottom diameter of 22 cm. To ensure drainage of effluents produced during ensiling, 3.5 kg of fine sand was placed at the bottom of each mini silo. A layer of cotton was placed on top of the sand layer to prevent direct contact between the forage and the sand. The ensiled mass amounted to 4.5 kg and was compacted using wooden stakes 1.5 m long and 10 cm in diameter to achieve a density of almost 600 kg/m³ [26], filling the remaining volume in the bucket.

The experimental silos were then closed with a lid, and the seal was reinforced with transparent adhesive tape stored in a closed room at room temperature (23 ± 3 °C) for 120 days. After 120 days of fermentation, the experimental silos were opened and about 5 cm of the top layer of each silo was discarded to avoid possible contamination [13]. The other part of the silage was homogenized and sampled for later analysis. This procedure ensured that representative samples were obtained from each treatment to assess silage quality.

2.3. Dry Matter Losses

The determination of total DM losses was calculated by determining the difference between the initial and final gross DM weight in the silos in relation to the amount of ensiled feed mass:

$$\mathrm{D}\mathrm{M}\mathrm{l}\mathrm{o}\mathrm{s}\mathrm{s}\mathrm{e}\mathrm{s}={\displaystyle \frac{(\mathrm{D}\mathrm{M}\mathrm{i}-\mathrm{D}\mathrm{M}\mathrm{f}).}{\mathrm{D}\mathrm{M}\mathrm{i}}}\times 100,$$

(1)

where DMlosses = total DM loss (%); DMi = initial DM amount (weight of silo after filling (kg)—weight of empty set, without forage, before filling (dry tare) (kg) × DM content of the forage in silage (% DM); DMf = final DM quantity (weight of the full silo before opening (kg)—weight of the empty set, without forage, after opening the silos (wet tare) (kg) × DM content of the forage at opening (% DM).

2.4. pH, Organic Acids, and Chemical Composition

To evaluate the pH and concentration of organic acids in the ensiled materials, an aqueous extract was prepared from 25 g of ground silage in an industrial blender with 225 mL of distilled water for one minute. This extract was then subjected to pH evaluation using a digital potentiometer (Tecnal, model Tec-3MP). The extract was then acidified 1:1 with H₂SO₄, diluted with distilled water and frozen for later analysis of the organic acids (lactic, acetic, and butyric acids).

To quantify the organic acids, the samples were treated with calcium hydroxide and copper sulfate and analyzed by HPLC (SPD-10 AVP, Shimadzu, Kyoto, Japan) according to the method described by [27]. The HPLC instrument was equipped with a refractive index detector and an Aminex HPX-87H column (BIO-RAD, Hercules, CA, USA), where the mobile phase contained 0.005 H₂SO₄ and had a flow rate of 0.7 mL/min at 45 °C.

Sampling of the fresh material was carried out before ensiling to determine the DM content. After opening, a new sampling was performed, the samples were partially dried at 55 °C for 72 h (method INCT-CA G-001/2) and then ground in a Willey-type knife mill with 2- and 1 mm sieves, respectively.

Samples ground to 1 mm were analyzed for DM (INCT-CA method G-003/1), crude protein (CP; INCT-CA method N-001/2), ether extract (EE; INCT-CA method G-005/2), and ash (INCT-CA method M-001/2) content [28]. In addition, the samples were analyzed for the content of neutral detergent fiber (NDF; INCT-CA method F-013/1), acid detergent fiber (ADF; INCT-CA method F-015/1), and lignin (INCT-CA method F-005/2) [28]. The samples ground to 2 mm were used to determine the indigestible neutral detergent fiber (NDFi) content (INCT-CA method F-009/2) as described by [28]. The samples were incubated in the rumen of two lactating Holstein cows through a rumen cannula for a period of 288 h, and the residual NDF was determined. Non-fibrous carbohydrate (NFC) was calculated using the equation proposed by [29]:

$$\mathrm{N}\mathrm{F}\mathrm{C}=100-(\mathrm{N}\mathrm{D}\mathrm{F}+\mathrm{C}\mathrm{P}+\mathrm{E}\mathrm{E}+\mathrm{A}\mathrm{s}\mathrm{h}),$$

(2)

where NFC = Non-fibrous carbohydrate (g/kg); NDF = Neutral detergent fiber (g/kg); CP = Crude protein (g/kg); EE = Ether extract (g/kg).

2.5. In Situ Rumen Degradability

Three rumen-fistulated lactating Holstein cows with an average body weight of 685 ± 119.1 kg and an average milk production of 37 ± 5.3 kg were used for the in situ incubation, which were kept in a compost barn system with ad libitum access to feed and water. The same ratio of corn silage to concentrate (45:55) was maintained throughout the assessment period of in situ rumen degradability and the animals were adapted for 14 days [30].

For incubation, samples were ground through a 2 mm sieve and approximately 4.5 g of each dried sample was individually weighed into nylon bags (16 × 10 cm, 50 μm pore size, Ankom Technology, Macedon, NY, USA). The samples were placed in the bags at a rate of 10 to 20 mg DM/cm² of the usable area of the bags [31]. The bags were placed in the rumen in the reverse order of the incubation times and removed at the same time. The incubation times used were 0, 3, 6, 12, 24, 30, 48, 72, 120, and 240 h [28]. The number of nylon bags used varied depending on the incubation time to ensure a sufficient number of residual samples after the process, which is influenced by the specific degradation rate of each feed. Therefore, four bags were used for the 72, 120, and 240 h incubation periods, and two bags were used for the remaining periods.

The samples were incubated in the rumen by attaching the nylon bags to a steel chain (90 × 2 cm²) [32] with a weight (300 g) at the end to ensure their immersion in the rumen contents. After removal, all bags were washed under running water to remove excess external residues and then washed in a washing machine (5 consecutive 1 min washes) [33]. The bags were then partially dried for 72 h at 55 °C in an oven with forced ventilation (INCT-CA method G-001/2) and then placed in an oven at 105 °C for 2 h to estimate the disappearance of DM in the rumen [32,33,34].

After drying and weighing the bags, the residue of each sample per treatment was removed from the nylon bags, ground in a Wiley mill (Fortinox, model STAR FT 50/6) with a 1 mm sieve and placed in a plastic bag labeled to obtain a sample from each animal and its respective incubation/treatment time. Subsequently, the residual samples were analyzed for DM and NDF content according to the methodology described by [28] and used to estimate the parameters of rumen-induced degradation of DM and NDF.

To estimate the in situ degradability of DM and NDF, the first-order asymptotic model proposed by Orskov and [35] was used, which utilizes the following equation:

$$\mathrm{D}\mathrm{e}\mathrm{g}\left(\mathrm{t}\right)=\mathrm{a}+\mathrm{b}\left(1-{\mathrm{e}}^{-\mathrm{c}\mathrm{t}}\right),$$

(3)

where Deg(t) = represents the degradability or disappearance of the nutrient (DM or NDF) from the feed, expressed as percentage; a = fraction of the feed that is soluble in water at time zero; b = fraction that is insoluble in water but potentially degradable in the rumen at a given time; c = rate of degradation of the potentially degradable fraction in the rumen (b); t = incubation time (hours).

The parameters a, b, and c of DM and NDF determined for each incubation period and each animal were organized for subsequent statistical analysis. In addition, the potentially degradable fraction and estimated degradation for passage rates of 2, 5, and 8%/h were calculated as described below:

$$\begin{gathered} \mathrm{P}\mathrm{D}=\mathrm{a}+\mathrm{b} \\ \mathrm{D}\mathrm{e}\mathrm{g}=\mathrm{a}+(\mathrm{b}\times \mathrm{c}/(\mathrm{c}+\mathrm{k}\mathrm{p}), \end{gathered}$$

(4)

where PD = represents the potentially degradable fraction; a = fraction of the feed that is soluble in water at time zero; b = fraction that is insoluble in water but potentially degradable in the rumen at a given time; Deg = represents the estimated degradable fraction according to the estimated parameters; c = degradation rate of the potentially degradable fraction in the rumen (b); kp = passage rate, which in this case can take a value of 2, 5, or 8%/h.

2.6. Statistical Analysis

Data related to the fermentation profile and chemical composition were subjected to analysis of variance using the GLIMMIX procedure of SAS [36] using the normal distribution for all variables analyzed. The data on the in situ degradation parameters were subjected to an analysis of variance using the lm package of R [37]. In all analysis, the F-test was used at a significant level of 5% according to the model below.

$${\mathrm{Y}}_{\mathrm{i}\mathrm{j}}=\mathrm{\mu }+{\mathrm{\alpha }}_{\mathrm{i}}+{\mathrm{\epsilon }}_{\mathrm{i}\mathrm{j}}$$

(5)

in which Yij is the measured response variable that received the ith treatment, μ is the general constant, αi is the fixed effect of the ith treatment, and εij is the random error term.

To compare the means between treatments, the Tukey test was used, and effects were considered significant at p ≤ 0.05 and trend at 0.05 < p ≤ 0.10.

3. Results

The mean values of pH, DM loss, and concentration of lactic, acetic, and butyric acids differed significantly among treatments (p < 0.01 for all variables; Table 3). The SGC and WI3 treatments had the lowest pH values, with 4.55 and 4.52, respectively. Additionally, SGC and WI3 treatments also had the lowest DM loss, with values of 1.69 and 1.64%, respectively. Lactic acid was higher in the SGC treatment, while the acetic acid had the lowest value in the CON treatment (Table 3). The butyric acid concentration was lower in the SGC and WI3 treatments, with 0.27 and 0.33%, respectively.

Table 3. pH, DM loss, and concentration of lactic, acetic, and butyric acid in Cenchrus purpureus (Schumach.) Morrone cv. BRS Capiaçu silage with addition of cornmeal and different wilting periods.

Parameters	Treatments 1				SEM 2	p-Value
	CON	SGC	WI3	WI5
pH	5.07 a	4.55 c	4.52 c	4.95 b	0.073	<0.01
DM Loss (%)	3.16 a	1.69 b	1.64 b	2.90 a	0.217	<0.01
Lactic Acid (% DM)	0.83 c	2.32 a	1.57 b	1.65 b	0.161	<0.01
Acetic Acid (% DM)	0.44 b	0.75 a	0.88 a	0.82 a	0.056	<0.01
Butyric Acid (% DM)	1.20 a	0.27 c	0.33 c	0.54 b	0.113	<0.01

¹ CON = control; SGC = additional 8% of cornmeal; WI3 = wilting for 3 days; WI5 = wilting for 5 days. ² SEM = standard error of the mean. Means followed by the same letter in the row do not differ from each other by Tukey’s test at 5% probability.

The contents of DM, ash, CP, EE, NDF, ADF, lignin, NFC, and NDFi were different among treatments (p < 0.01; Table 4). Treatments WI3 and WI5 presented the greatest DM values with 318.3 and 305.4 g/kg, respectively. Treatment WI5 presented the lowest CP value among treatments, with 51.4 g/kg, while the other treatments average 60.2 g/kg. Treatments CON and SGC had the greatest values of EE. When evaluating the NDF, ADF, and NDFi contents, the SGC treatment had the lowest values (700.2 g/kg, 453.3 g/kg, and 355.0 g/kg, respectively). The SGC and WI3 treatments had the lowest values of lignin, with 82.1 and 88.4 g/kg, respectively. The greatest value of NFC was observed in SGC treatment (138.6 g/kg), while the other treatments averaged 61.3 g/kg (Table 4).

Table 4. Chemical composition of Cenchrus purpureus (Schumach.) Morrone cv. BRS Capiaçu silage with the addition of cornmeal and different wilting periods.

Items (g/kg) 1	Treatments 2				SEM 3	p-Value
	CON	SGC	WI3	WI5
Dry Matter	223.1 c	257.2 b	318.3 a	305.4 a	4.50	<0.01
Ash	80.2 b	80.5 b	90.6 a	85.5 ab	1.59	<0.01
Crude Protein	58.4 a	63.8 a	58.5 a	51.4 b	1.35	<0.01
Ether Extract	16.7 a	16.7 a	9.9 b	7.5 b	1.05	<0.01
NDF	792.7 ab	700.2 c	765.8 b	798.2 a	6.45	<0.01
ADF	523.2 ab	453.3 c	504.0 b	540.5 a	7.26	<0.01
Lignin	97.4 ab	82.1 c	88.4 bc	101.1 a	2.08	<0.01
NFC	51.7 b	138.6 a	74.9 b	57.2 b	6.36	<0.01
NDFi	420.4 a	355.0 b	394.5 a	414.9 a	7.49	<0.01

¹ NDF = neutral detergent fiber; ADF = acid detergent fiber; NFC = non-fibrous carbohydrates; NDFi = indigestible neutral detergent fiber. ² CON = control; SGC = additional 8% of cornmeal; WI3 = wilting for 3 days; WI5 = wilting for 5 days. ³ SEM = standard error of the mean. Means followed by the same letter in the row do not differ from each other by Tukey’s test at 5% probability.

For the in situ rumen degradability evaluation, only the fraction “a” of DM (p < 0.01) and the effective degradability of DM at passage rates of 5 and 8%/h (p = 0.01) showed significant differences between treatments and a tendency towards the rate of 2%/h (Table 5 and Table 6). The fraction “a” of DM of the SGC treatment was 289 g/kg, while the other treatments averaged 227 g/kg, which was about 21.2% lower (Table 5). Treatments WI3 and WI5 showed greater values for the fraction “a” of DM when compared to the CON treatment (Table 5). No differences (p > 0.05) were observed among treatments in terms of the potentially degradable fraction in the rumen (fraction b) and the rate of degradation of the potentially degradable fraction in the rumen (c) of the DM. Similarly, there were no significant differences (p > 0.05) among treatments for the degradation of fractions “a”, “b” and the degradation rate of NDF (Table 5).

Table 5. In situ degradation parameters of DM and NDF of Cenchrus purpureus (Schumach.) Morrone cv. BRS Capiaçu silage with the addition of cornmeal and different wilting periods.

Items	Treatments 1				SEM 2	p-Value
	CON	SGC	WI3	WI5
DM parameters
a, g/kg	209 c	289 a	240 b	234 b	4.95	< 0.01
b, g/kg	336	330	332	314	26.7	0.93
c, %/h	2.03	2.27	1.85	1.76	0.21	0.39
NDF parameters
a, g/kg	149	162	160	173	7.63	0.27
b, g/kg	370	382	368	347	30.6	0.88
c, %/h	1.87	2.11	1.80	1.77	0.19	0.63
Potential degradation of DM, g/kg	545	619	572	548	25.6	0.22
Potential degradation of NDF, g/kg	519	544	528	520	31.6	0.93

¹ CON = control; SGC = additional 8% of cornmeal; WI3 = wilting for 3 days; WI5 = wilting for 5 days. ² SEM = standard error of the mean. Means followed by the same letter in the row do not differ from each other by Tukey’s test at 5% probability.

Table 6. Effective degradability of DM and NDF at three passage rates of Cenchrus purpureus (Schumach.) Morrone cv. BRS Capiaçu silage with the addition of cornmeal and different wilting periods.

Items	Treatments 1				SEM 2	p-Value
	CON	SGC	WI3	WI5
Effective degradability of DM (g/kg)
Kp = 2%/h	379 B	465 A	400 AB	381 B	20.7	0.06
Kp = 5%/h	307 b	393 a	330 b	316 b	13.6	0.01
Kp = 8%/h	278 b	362 a	303 b	291 b	10.2	0.01
Effective degradability of NDF (g/kg)
Kp = 2%/h	329	359	335	336	25.1	0.85
Kp = 5%/h	251	276	258	264	17.3	0.77
Kp = 8%/h	221	243	228	236	13.8	0.70

¹ CON = control; SGC = additional 8% of cornmeal; WI3 = wilting for 3 days; WI5 = wilting for 5 days. ² SEM = standard error of the mean. Means followed by the same lowercase in the row do not differ from each other by Tukey’s test at 5% probability, while means followed by the same uppercase in the row do not differ from each other by Tukey’s test at 10% probability.

No significant differences (p > 0.05; Table 5) were found between the treatments regarding the potential degradation of DM and NDF. The effective degradability of DM at rates of 5% and 8%/h in the SGC treatment was greater than in the other treatments (Table 6).

4. Discussion

The material was harvested with 22% DM, which is below the range recommended by [25] (25–35%) for the anaerobic fermentation of grasses at the time of ensiling. Tropical grasses with a DM content lower than 25% have a higher potential for nutrient losses as gases and effluents, which can affect the nutritional quality of the silage, as some of the carbohydrates and proteins are lost in the effluents [25,26,27,28,29,30,31,32,33,34,35,36,37,38].

The lower pH values found in the SGC and WI3 treatments are consistent with those of [25], who prepared a review on the interpretation of the chemical, microbial, and organoleptic components of different types of silage. For grass silage, prior literature indicates that at a DM content of 25–35%, the optimal pH lies between 4.3 and 4.7. Under these benchmarks, the SGC and WI3 treatments consistently exhibited pH values that fell within the recommended interval, indicating superior fermentation quality relative to established guidelines.

The pH values observed in SGC and WI3 likely stem from reduced grass moisture, which concentrated soluble carbohydrates and thereby increased the silage’s osmolarity. These factors probably promoted a reduction in buffering capacity and exacerbated the drop in pH [39]. Lowering the pH helps to prevent the growth of spoilage microorganisms such as yeasts, molds, and undesirable aerobic bacteria, thus helping to preserve the nutritional value of the ensiled material [40].

The lower DM loss in the SGC and WI3 treatments could be related to the DM content of these treatments at the ensiling time. Tropical forages with a dry matter content lower than 25% may experience significant nutrient losses due to the release of gases and/or effluents, which negatively affect the nutritional value of the silages, as some of the carbohydrates and proteins present in the forages are released in the effluents during the fermentation process [25].

Amaral et al. [41] characterized the population of lactic acid bacteria in the fermentation of elephant grass and selected promising lines for inoculants to evaluate their effect on silages of the BRS Capiaçu variety harvested at 22% DM. These authors found DM losses between 3.8 and 12.3% in the different treatments. In the present study, lower DM losses were observed in the SGC and WI3 treatments, possibly due to the greater DM concentration at harvest, the addition of cornmeal, and the wilting process. Our results are consistent with those of [13], who showed that the addition of ground cornmeal minimizes losses and suggest that adjustments to the ensiling material and substrate conditions can optimize lactic acid production, regardless of the initial lactic acid bacteria population.

According to Bates et al. [42], DM loss in wilted silages may be increased due to less intensive fermentation characterized by lower acid production compared to non-wilted silages or silages that have been wilted for a shorter time. Our results confirm this observation and show that when comparing treatments WI3 and WI5, longer wilting resulted in greater DM loss.

The greatest lactic acid concentrations were observed in treatments with greater DM content, which underlines the expected relationship between greater lactic acid production and greater DM content. In addition, the increase in lactic acid production could be due to the fact that the ensiled material had a greater amount of non-fibrous carbohydrates due to the addition of cornmeal, providing more fermentable carbohydrates for lactic bacteria [10,43,44].

Kung et al. [25], considered the ideal lactic acid concentration ranging from 6 to 10% for grass silage with a DM content of 25 to 35%. Therefore, none of our treatments had concentrations within this ideal range. This low lactic acid concentration may be directly related to the lower concentration of water-soluble carbohydrates (WSCs) found in tropical forages. As WSC are the main substrate utilized by the lactic acid bacteria during the silage fermentation process [45]. During wilting, enzymatic respiration occurs, which leads to the loss of WSC [46], which also explains the lower concentration of lactic acid in treatments WI3 and WI5 compared to SGC. When comparing the CON treatment with WI3 and WI5, we found that the treatments that underwent the wilting process had greater lactic acid levels. References [42,43,44,45,46,47] suggest that wilting lowers silage water activity, hindering Clostridium proliferation during fermentation. This condition favors lactic acid bacteria and affords them more time to generate adequate organic acids, which are crucial for effective silage preservation. However, lower lactic acid likely stems not only from low WSC but also from higher DM after pre-wilting, the strong buffering of C4 grasses, a more heterofermentative microbiota, and management/fermentation factors (packing, O₂, temperature, time).

According to Santos et al. and Ferreira et al. [48,49], greater lactic acid production can lead to lower DM losses in grass silages. This occurs because lactic acid fermentation causes only minimal losses, whereas acetic and butyric fermentation are associated with secondary processes and DM losses as gaseous emissions. Regarding acetic acid concentration, the values in all treatments were below the reference value proposed by [25] for grass silages with a DM content between 25 and 35%, with the ideal range being 1 to 3% of DM. In contrast, the values found are greater than those observed by [50] in wilted elephant grass silages or with the addition of cassava bran, which had an average of 0.47% acetic acid. Acetic acid plays an important role in the preservation of silage, especially after opening the silo, as it inhibits filamentous fungi, moulds, and yeasts in the silage and thus contributes to greater aerobic stability [51].

It is worth mentioning that acetic acid also contributes to the preservation of ensiled material. However, greater concentrations indicate that undesirable changes may occur in the ensiling process. The ideal fermentation should take place with a greater amount of lactic acid to ensure a better quality of the fermented material [52]. The lower concentration of butyric acid in the silage could be related to the decrease in pH, which may have inhibited the growth and proteolytic activity of microorganisms such as Clostridia [53,54]. When considering the butyric acid concentrations proposed by [25], only the CON treatment showed production above the maximum limit of 1% recommended as a quality parameter for this type of silage.

As described by Jobim et al. [55], the ash content in silages can indicate losses of organic compounds due to contamination of the forage with sand during cutting, which normally occurs in the ensiling process. These losses lead to an increase in ash content and can have a negative effect on the pH buffering capacity of the silage. In addition, this increase in ash content may also be related to a dilution effect due to the lower water concentration of these treatments compared to the control, resulting in less dilution of this component.

Regarding CP, a similar result was observed by [14,15,16,17,18,19,20,21,22,23,24,25,26,27,28,29,30,31,32,33,34,35,36,37,38,39,40,41,42,43,44,45,46,47,48,49,50,51,52,53,54,55,56]. Paula et al. [14], investigated the bromatological composition of BRS Capiaçu elephant grass silage with cornmeal added in proportions of 0, 5, 10, 15, and 20%. The authors found that the addition of ground corn to the silages had a linear effect, resulting in higher CP levels, as the cornmeal contained a greater proportion of this component compared to the elephant grass cv. BRS Capiaçu. In this study, the addition of 10% of cornmeal led to an increase of 0.93 percentage points in the CP content of the DM, while in our study the addition of 8% cornmeal led to an increase of 0.54 percentage points in the CP content of the DM. However, the SGC treatment CP was similar to CON and WI3 treatments. The lower CP content in the WI5 treatment can be attributed to the prolonged wilting time. According to [57], a wilting period of several days can result in a loss of water-soluble carbohydrates and a reduction in proteins due to the deamination of amino acids, which can lead to increased ammonia-nitrogen production during silage fermentation.

A similar result was observed by [14] regarding the EE content, who found that despite the numerical increases in EE levels in the silage with the inclusion of ground cornmeal, there was no significant difference compared to the control treatment. The lowest EE content was observed in treatments WI3 and WI5, which can be attributed to wilting time. Studies suggest that wilting the material before ensiling reduces the EE content, resulting in significant losses of specific fatty acids, such as α-linolenic acid and linoleic acid [58,59,60]. These losses are associated with the lipoxygenase system, a plant defense mechanism that is activated in damaged tissues [59].

The addition of 8% cornmeal in the SGC treatment decreased the fibrous fractions, with a reduction of 1.16%, 0.87%, and 0.82% for NDF, ADF, and NDFi, respectively, for each 1% of cornmeal added. These results can be explained by a dilution effect, as cornmeal contains lower levels of these fractions compared to BRS Capiaçu grass [14]. In contrast, the greater levels of these components observed in treatments CON, WI3, and WI5 may be related to the loss of WSC during silage fermentation [61]. The NDF and ADF content is an indicator of the quantity and quality of fiber contained in the material, as a high NDF content may limit intake, while a low ADF content indicates fiber of good quality [62].

When evaluating the use of citrus pulp and wilting on the quality of elephant grass silage, ref. [63] found that citrus pulp reduced the lignin content due to the dilution effect. On the other hand, wilting increased the lignin content, a change that was not observed in the WI3 treatment of the present study. In contrast, refs. [16,17,18,19,20,21,22,23,24,25,26,27,28,29,30,31,32,33,34,35,36,37,38,39,40,41,42,43,44,45,46,47,48,49,50,51,52,53,54,55,56,57,58,59,60,61,62,63,64] found no significant differences between wilted and non-wilted grasses. As plant tissue matures, secondary cell wall thickness increases, leading to an increase in NDF and lignin concentration at the expense of cell content [62,65,66].

When evaluating the rumen degradability, the fraction “a” of DM corresponds to the proportion of the feed that is readily available for the action of the microorganisms in the rumen. This increase in fraction “a” can be attributed to the inclusion of ground cornmeal, an ingredient with high values of the water-soluble fraction in its composition [67]. The fraction “a” of DM for SGC treatment was 289 g/kg, a result similar to the found by [67], who evaluated the in situ rumen degradability of elephant grass silages supplemented with corn bran and autochthonous microbiota inoculant using 60-day-old grass. These authors observed higher values of fraction “a” in the grass silage with the addition of ground cornmeal, with average values of 273 and 257 g/kg for the treatments without and with inoculant, respectively.

The values of the fraction “a” of DM for WI3 and WI5 treatments did not agree with the results of [16], who found no significant effects of wilting times on the rumen kinetics parameters of DM. The authors attributed this pattern to the decrease in CP and EE content with increasing wilting time. However, this pattern was not observed in the present study, because despite the reduction in CP and EE contents, the WI3 and WI5 treatments still showed greater values for the fraction “a” of DM compared to the CON treatment. This pattern could be due to the greater DM content in both treatments. According to [68], in their meta-analysis on the effects of wilting on silage quality, the authors concluded that during wilting, the concentration of carbohydrates, including WSC, increases as the forage loses moisture. In addition, wilting slows down the respiration of plant cells, which leads to a reduction in carbohydrate consumption. This reduction in respiration favors the conservation of carbohydrates, including WSC, which contributes to a greater content of soluble fractions in the forage [10].

Regarding the NDF degradation rate, it is important to emphasize that according to [69], the NDF degradation rate must be between 0.02 and 0.06%/h for bulky feeds to be considered of high quality, and only the SGC treatment was within this recommended range, which highlights that the treatments did not change fiber structure.

Ribas et al. [16] reported an average potential DM degradability of 515.9 g/kg for BRS Capiaçu grass silage, a value that the authors classified as low and possibly related to the high iNDF content. This result is close to the average value of 571 g/kg found in our study. The greater effective degradability of the DM in the SGC treatment can be attributed to the better degradability of corn compared to elephant grass, as it has easily soluble components compared to the BRS Capiaçu variety [70].

It is worth noting that although the wilting of the forage was performed for 3 days and 5 days during the winter, the time required to reach the desired dry matter content may vary depending on local climatic conditions. In warmer regions, with higher average temperatures, fewer days may be necessary to achieve the same dry matter. Therefore, it is important to consider other factors such as the dry matter content of the plant, local climate, and environmental conditions when adapting the wilting time for different regions, thus ensuring silage quality without compromising the efficiency of the process.

5. Conclusions

The wilting of BRS Capiaçu elephant grass for 3 days before ensiling or the addition of cornmeal with the aim to increasing the DM content are strategies that improve the fermentation quality of the ensiled material and enable better forage conservation. However, only the SGC treatment showed better degradation of dry matter. It is important to highlight that wilting is an advantageous technique, especially in situations where family labor is used, since the cost is lower, and the results of silage quality are similar to those obtained with the addition of cornmeal. However, it is worth noting that, as mentioned earlier, the results depend on the wilting time, which must be adjusted according to the specific conditions of the plant and climate. On the other hand, the addition of moisture sequestrants can improve silage quality without major issues caused by rainfall in the harvest. However, this can also significantly increase the cost of silage production.