Aerobic deterioration of silage: processes and prevention

Efficient preservation of forages as silage requires minimizing losses during the aerobic, fermentation, storage and feedout phases. The main concern about silage losses is in cases where losses are large (20% or more); when silages are fed in warm, humid weather; when horizontal silos have a large surface area and feedout face; and in silages where fermentation was restricted (Ohyama et al., 1975; McGechan, 1990). While quality of the fermentation phase has in general been improved over the past years, the same cannot be said about aerobic stability of silages in the feedout phase (Honig et al., 1999). The improvements in silage fermentation quality, which prevent butyric acid and minimize the amount of acetic acid, have increased the risk of aerobically unstable silages (Wyss, 1999). This is why well-preserved silages are often considered to be more prone to aerobic deterioration than poorly fermented silages (Cai et al., 1999).

All silages when exposed to air sooner or later deteriorate as a result of aerobic microbial activity (Petterson, 1988; Jonsson, 1989; Ashbell and Weinberg, 1992). This inevitable aerobic deterioration usually results in high dry matter (DM) loss (Woolford, 1990; Bolsen, 1997) and the loss of important nutritional components (Kung et al., 1998) by the oxidation of lactic acid and water soluble carbohydrates (WSC), which also leads to reduced preservation quality (Cai et al., 1999; Holzer, et al., 1999). The accumulation of degradation products can affect palatability and cause feed refusal by livestock (Lindgren et al., 1988; Holzer et al., 1999). Some aerobic microorganisms such as molds, bacilli, and Listeria monocytogenes can be harmful to livestock (Lättemäe, 1997; Kautz, 1998; Driehuis et al., 1999); and further aerobic deterioration can result in the formation of mycotoxins, which can be potentially lethal (Holzer et al., 1999). The prevention of aerobic deterioration of silage is an important part of every silage program (Cai et al., 1999).

Efficient preservation of forages as silage requires minimizing losses during the aerobic, fermentation, storage and feedout phases. The main concern about silage losses is in cases where losses are large (20% or more); when silages are fed in warm, humid weather; when horizontal silos have a large surface area and feedout face; and in silages where fermentation was restricted (Ohyama et al., 1975; McGechan, 1990).

While quality of the fermentation phase has in general been improved over the past years, the same cannot be said about aerobic stability of silages in the feedout phase (Honig et al., 1999). The improvements in silage fermentation quality, which prevent butyric acid and minimize the amount of acetic acid, have increased the risk of aerobically unstable silages (Wyss, 1999). This is why well-preserved silages are often considered to be more prone to aerobic deterioration than poorly fermented silages (Cai et al., 1999).

All silages when exposed to air sooner or later deteriorate as a result of aerobic microbial activity (Petterson, 1988; Jonsson, 1989; Ashbell and Weinberg, 1992). This inevitable aerobic deterioration usually results in high dry matter (DM) loss (Woolford, 1990; Bolsen, 1997) and the loss of important nutritional components (Kung et al., 1998) by the oxidation of lactic acid and water soluble carbohydrates (WSC), which also leads to reduced preservation quality (Cai et al., 1999; Holzer, et al., 1999). The accumulation of degradation products can affect palatability and cause feed refusal by livestock (Lindgren et al., 1988; Holzer et al., 1999). Some aerobic microorganisms such as molds, bacilli, and Listeria monocytogenes can be harmful to livestock (Lättemäe, 1997; Kautz, 1998; Driehuis et al., 1999); and further aerobic deterioration can result in the formation of mycotoxins, which can be potentially lethal (Holzer et al., 1999). The prevention of aerobic deterioration of silage is an important part of every silage program (Cai et al., 1999).

The microbial processes involved in aerobic spoilage have been well established. Most research suggests that fungi, particularly yeasts, have a The microbial processes involved in aerobic spoilage have been well established. Most research suggests that fungi, particularly yeasts, have a Factors involved in aerobic deterioration The aerobic deterioration of silage is a complex process (McGechan, 1990). Microbial agents as well as ensiling practices are factors that contribute to losses. The factors influencing deterioration include oxygen (and exposure time), composition of the microbial population, substrate type (e.g., nutrient content of the silage), stage of maturity at harvest, density of the silage, ambient temperature and temperature of the silage mass (Ohyama et al., 1975; Woolford, 1978; McGechan, 1990).


SILAGE pH

The occurrence of aerobic deterioration in a silage can be monitored by measuring pH. Unless the original silage pH is high (ca >5.0), the initial deterioration processes result in an increase in pH (Ohyama et al., 1975).

Under anaerobic conditions, the low pH of a silage inhibits the growth of undesirable microorganisms (e.g., clostridia). However, it has been shown that low pH per se is not sufficient to prevent aerobic deterioration, because yeasts can grow under fairly low pH conditions (Ohyama et al., 1975).


DRY MATTER CONTENT

It has been assumed that aerobic deterioration would take place more frequently in drier forages with low packing densities, which would allow air to invade the silage rather easily. However, the results of Ohyama et al.

(1975) did not support this assumption. Pessi and Nousiainen (1999) reported that well fermented silages with low DM content and slightly pre-wilted silages that were preserved with or without a bacterial inoculant appeared to be prone to aerobic spoilage. Pitt et al. (1991) observed that high DM silages were less stable during the feedout phase because they were associated with a high pH and low acid concentrations. Furthermore, the trend toward increasing the extent of pre-wilting has made it more difficult to consolidate forages and prevent the exchange of gases (i.e., allowing air infiltrate the silage mass).

Data from Wyss (1999) showed that grass silages with a higher degree of pre-wilting were more aerobically unstable than grass silages with a lower DM content. The author concluded that the higher sugar content, lower concentration of acetic acid (from a less intensive fermentation), a higher population of yeasts and lower consolidation in the pre-wilted silages compared to their low DM counterparts contributed to the differences in aerobic stability.


RESIDUAL WATER SOLUBLE CARBOHYDRATE CONTENT

An increase in the temperature of the silage mass is another useful criterion signaling that silage is undergoing aerobic deterioration (Ohyama et al., 1975). Heating is caused by the metabolism of sugars and organic acids by yeasts and bacteria (Spoelstra et al., 1988). The concentrations of lactic acid and WSC usually decrease very rapidly as they are used as substrate by the aerobic microorganisms (Ohyama et al., 1975).

Because fungal growth is more rapid on WSC, stability decreases when high residual WSC levels are present (Pitt et al., 1991). Ohyama et al. (1975) reported cases where silages remained stable in spite of high levels of WSC. However, once deterioration occurred, losses were larger in silages with high WSC contents.


TYPE OF CROP ENSILED

The influence of forage type on aerobic stability is indirect through differences in fermentability of the crop (Woolford, 1984). Whole-plant corn and small grain cereal silages are generally more prone to aerobic deterioration than grass or legume silages (Kautz, 1998).

Aerobic spoilage of corn silage is a well-documented problem, which generally occurs through a series of fungal and bacterial interactions (Oude Elferink et al., 1999; Pahlow et al., 1999). Since corn has an adequate reserve of readily available fermentable sugars, it ensiles without difficulty and achieves a low, stable pH quite rapidly. The high ‘ensileability’ of corn is attributed to substrate availability and also to the relatively low content of organic acids, which buffer against acidification (Woolford, 1984).


TEMPERATURE

Although it is well documented that higher temperatures (e.g., 30-45°C) generally favor the types of microbial activity that lead to rapid aerobic deterioration, it is important to note that deterioration can occur even at ambient temperatures of 10-15°C. However, some silages are aerobically stable even at higher temperatures (Pitt et al., 1991). Ohyama et al. (1975) observed that silages that were stable at 25-30°C contained relatively high concentrations of propionic acid (0.28-0.46% of the DM) or butyric acid (0.56-0.92% of the DM).


DENSITY

Spoilage due to air infiltration depends on the rate of gas movement in silage, and it is related to factors such as porosity, density, stage of maturity (e.g., more mature forage contains a higher proportion of fiber and is more rigid) and chop length. A high density in the ensiled forage is important since density and DM content determine the porosity of the silage. Porosity, in turn, will set the rate at which air will move into the silage mass (McGechan, 1990; Muck and Holmes, 1999). Research into the mechanism of air infiltration into silage suggests that shorter chopped forages can offer less resistance to gas movement than longer chopped material, despite a higher density in the shorter forage (McGechan, 1990).

Ensuring sufficient consolidation is important, especially in view of the continued increasing harvest capacities (tons of DM harvested per hour). Depth of air penetration is largely determined by harvesting, filling, storing and feedout practices. In general, as density decreases, the depth of air ingress increases. The progression of the feedout face through the silo is important, and is dependent on the height and width of the silo, herd size, and the amount of silage fed per day (Honig et al., 1999). Lindgren et al. (1988) reported a reduction in the populations of aerobic spoilage microbes as silage depth increased.

Diverse flora of yeasts, bacilli and Enterobacteriaceae have been observed in the course of experiments; and because the frequency of the different species varies from year to year and from silage to silage, it is impossible to generalize regarding microbial dynamics during fermentation, storage and feedout phases (Lindgren et al., 1985). However, there appears to be a very restricted range of yeast and bacterial genera involved in the aerobic deterioration of widely differing silages (Woolford, 1990).

The dynamics probably depend on the establishment and/or survival of the aerobic spoilers during the fermentation and storage phases. Little is known about the factors that influence their establishment. Oxygen diffusion during storage appears to be important for the establishment of lactateassimilative yeasts. A 0.1 mm polyethylene sheet did not prevent the diffusion of oxygen into bunker silages (Lindgren et al., 1985).


YEASTS

Most yeasts grow well at temperatures between 0ºC and 37ºC, and few are adapted to temperatures above 45ºC (Davenport, 1980). Jonsson (1989) reported that yeast counts decreased as the temperature during aerobic deterioration exceeded 40ºC. Yeasts are more sensitive to higher temperatures than are clostridia (Silliker et al., 1981 as cited by Jonsson, 1989). Yeast numbers are often found to increase during wilting (Henderson et al., 1972; Woolford, 1984; Jonsson and Pahlow, 1989; Woolford, 1990). This is particularly evident for lactate-utilizing and fermentative yeasts, which were below the limit of detection at the time of harvest (about 200 CFU/g of DM), increasing to more than 10,000 CFU/g of DM during the wilting process. This explains why it was assumed that aerobic deterioration occurred more frequently with drier forages that were ensiled at low densities (Ohyama et al., 1975). Soil contamination (e.g., during mechanical tedding) likely contributes to the increased yeast counts, because soil is a reservoir from which yeasts are transferred to ensiled crops (Jonsson, 1989).

The low pH of most silages does not inhibit the survival of most yeasts, as yeasts can grow within a pH range of 3-8 (Woolford, 1976). The optimum pH for the growth of most species is 3.5-6.5. Under aerobic conditions yeast tolerate organic acids better than most other microorganisms (Woolford, 1975a; Silliker et al., 1980 as cited by Jonsson, 1989). Yeasts isolated from silage consume organic compounds such as lactic, acetic, citric, malic, succinic and propionic acids and ethanol under aerobic, but not under anaerobic conditions (Middlehoven and Franzen, 1986). In studies reported by Henderson et al. (1972) and Jonsson (1989) yeasts were favored by the addition of formic acid because of increased amounts of residual sugars that contributed to their survival.

The majority of epiphytic yeasts on forage crops are non-fermentative yeasts of the genera Sporobolomyces, Cryptococcus, Rhodotorula and Torulopsis, which can range in population density from less than 10 to 106 or 107 CFU/g of fresh material (McDonald, 1981; Lindgren et al., 1985; Middlehoven and van Baalen, 1988).

After anaerobiosis has been established in the ensiled forage, aerobic fungi are succeeded by a fermentative flora of yeasts (Di Menna et al., 1981; Middlehoven and van Baalen, 1988). The dominate species are typically C. lambica, C. krusei, H. anomala, Torulopis spp. and Saccharomyces spp. These species have also been found in other fermented products (Reed 1983, as cited by Jonsson, 1989). The varieties that develop under fermentation conditions are C. lambica and C. krusei. Besides lactic acid, there are only a few carbohydrates that these species can assimilate. In addition, C. lambica, together with Hansenula anomala, belong to the class of yeasts that assimilate xylose. This property becomes especially important in later stages of the fermentation phase of the ensiling process when the readily available sugars are depleted and xylose is available as the result of acid hydrolysis or the action of natural-occurring hemicellulases (McDonald, 1981). The yeast species found during this phase have been isolated from aerobically unstable silages by several researchers (Di Menna et al., 1981).

The presence of air during ensiling promotes the development of a different population of yeasts. Basically, the yeasts involved in aerobic deterioration have been classified into two groups: those that can use acids (Candida, Endomycopsis, Hansenula and Pichia) and those that use sugars (Torulopsis) (Ohyama et al., 1975; Middlehoven and Franzen, 1986; Middlehoven and van Baalen, 1988; Pettersson, 1988; Woolford, 1990).

Yeasts such as Candida and Hansenula species that are able to metabolize lactic acid are responsible for the increased pH of a silage, which creates conditions suitable for the growth of molds during an extended feedout phase. Yeasts can inhibit the production of lactic acid. Yeasts do not contribute to the preservation of an ensiled forage and actually compete with lactic acid bacteria for available WSC (Ruxton and McDonald, 1974).

Yeast numbers vary from less than 10 CFU/g on a fresh basis in wellpreserved silage to 1012 CFU/g on a fresh basis in aerobically deteriorated silage (Di Menna et al. 1981; Lindgren et al., 1985; Middlehoven and van Baalen, 1988; Pettersson, 1988). Daniel et al. (1970 as cited by Spoelstra et al., 1988) stated that silages with at least 100,000 yeasts per gram were very susceptible to aerobic spoilage. This critical value for yeast numbers is valid on the condition that the population of 105 yeasts per g of DM is made up principally of lactate-utilizing organisms (Jonsson, 1989).

The microflora on whole-plant corn is usually characterized by a high number of microorganisms, particularly yeasts, which might be 100 to 1000- fold higher than numbers observed on grasses or legumes. It has been suggested that because yeasts on corn account for a substantial proportion of the microflora, they play a major role in the ensiling process of this crop (Woolford, 1984). The predominant yeast flora of 13 different corn silages was shown to consist of Candida lambica, Saccharomyces dairensis, Saccharomyces exiguus, Candida holmii, Candida milleri and Candida krusei (Middelhoven and van Baalen, 1988). All of these species, except S. dairensis, tolerated acetic acid at pH 4.0. Lactic and acetic acids and ethanol were readily assimilated at pH 4.0 in the presence of yeast extract (1 g/L).


MOLDS

Oxygen is vital for the growth of molds, so silage management practices that eliminate the presence of air will prevent molds from having an active role in the ensiling process (Schlatter and Smith, 1999). The conditions that favor mold growth include: 1) a moisture content above 13%, 2) relative humidity above 70%, 3) temperature above 55ºF, 4) readily available nutrients, 5) pH greater than 5, and 6) the presence of oxygen (Gotlieb, 1997; Kautz, 1998).

Mold development in silage is inhibited by low pH but is encouraged by the presence of unfermented sugars. Simply stated, mold growth occurs when there is adequate moisture, warmth, and air. Growth of these fungi increases the pH of the ensiled material, which creates ideal conditions for other microorganisms to grow. This development of molds is typical of silage where anaerobic conditions are not maintained. Thus, contamination of silage can be extensive near the surface of the ensiled forage, in poorly sealed silos, in wrapped-bale silages that are not sealed properly, where polyethylene is damaged or in silages with a low packing density (Clarke, 1988).

Whole-plant corn, sorghum, or small grain cereals are more prone to mold growth than grasses or legumes. Molds include a wide range of genera (Fusarium, Aspergillus, Mucor, Penicillium, Monilla). Some molds proliferate while the crop is growing in the field prior to harvest while others propagate during the storage phase (Kautz, 1998). There are some thermophilic molds, but because they generally grow slower than yeasts, these molds are considered incidental and of little influence in the aerobic deterioration of silage (Woolford, 1990).

Penicillium rockefortii, which is suspected of suppressing the immune response in cattle, was recently isolated in silage that was within 25 cm of the surface in horizontal silos (Sundberg and Häggblom, 1999). This mold is also capable of growing at a low pH and in a nearly anaerobic environment (Detmer et al., 1999).


BACTERIA

While the majority of evidence favors fungi and yeasts, in particular, as being responsible for aerobic deterioration in silage, results of other investigations on the microbiological, physical and chemical changes that occur during aerobic deterioration suggest that bacteria initiate the process in silage (Woolford et al., 1982).

Results from Spoelstra et al. (1988) suggest that aerobic deterioration is often started by a simultaneous build-up of populations of acetic acid producing bacteria and yeasts. However, when silage was inoculated with acetic acid bacteria (from the genus Acetobacter), no growth of yeasts was observed, suggesting that acetic acid bacteria can be solely responsible for initiating aerobic deterioration. The primary substrate for these bacteria is ethanol, followed by lactic and acetic acids (Woolford, 1990).

Barry et al. (1980) attributed the onset of heating in the feedout phase to acid-tolerant aerobic bacteria. The thermotolerant Bacillus spp. can gradually replace the less thermotolerant yeasts as the temperature of the silage increases during the deterioration process. In addition, the organisms identified in silages (e.g., Bacillus spp. and Monascus ruber) are able to effectively utilize lactic acid. Monascus spp. have previously been found in aerobically deteriorated corn silage (Pelhate, 1977, and Hara and Ohyama (1979) (as cited by Jonsson, 1989) and Bacillus spp. are considered to be a cause of aerobic deterioration in corn silage (Woolford, 1978;1984).

The spoilage process can be initiated in corn silage by aerobic bacteria (such as Bacillus) followed by an increase in yeast populations (Jonsson, 1989). The reverse is generally true in legume, grass and cereal silages. Woolford (1978) (as cited by Spoelstra et al., 1988) and Woolford and Cook (1978) found that an antimycotic agent did not prevent aerobic deterioration in corn silage.

There are other bacteria, such as Clostridium and Listeria, that are a hazard to health and the hygienic quality of silage and its nutritional value (Woolford, 1990; Lättemäe, 1997). So far only seven species have been identified to play a role in the fermentation phase (C. butyricum, C. tyrobutyricum, C. paraputrificum, C. sporogenes, C. bifermentans, C. perfringens and C. sphenoides). Clostridia can be found in silage primarily because of soil contamination, and they can be either saccharolytic or proteolytic and can consume lactic acid, which causes a rise in silage pH. Clostridial growth in silage is favored by low DM content, low WSC content, high buffering capacity of the forage and high ambient temperature (>25- 30°C) (Lättemäe, 1997). The pH level below which clostridia stop growing decreases with decreasing DM content of the ensiled forage. As with the clostridia that are of significance to silage per se, a high DM and/or low pH can limit the growth of C. botulinum, however, its presence has been reported in silage that has undergone extensive secondary fermentation (Woolford, 1990). Clostridia, as well as Bacillus, can form spores that can survive pasteurization (Pettersson, 1988).

Listeria is commonly found in the soil and on decaying forage. Many strains of most species of Listeria have been isolated from silage, particularly near the surface and especially from low quality silages. Its growth is markedly decreased in acid medium with a critical pH value of 5.5 (Woolford, 1990). The effect of additives on aerobic deterioration

An ideal fermentation reduces fermentation losses and maintains an acceptable degree of aerobic stability during storage and feedout. Proper management and effective silage inoculants play a key role in balancing these two important silage traits (Kautz, 1998).

As mentioned previously, often the most poorly-fermented silages with a high content of NPN, butyric acid, or acetic acid are generally the most aerobically stable (Kautz, 1998) while the well preserved silages are the most prone to aerobic deterioration (Woolford, 1990). The two primary reasons for using an additive to improve aerobic stability are: 1) to prevent heating and DM losses, and 2) to prevent a reduction in livestock performance associated with feeding spoiled silage. An efficacious silage preservative should prevent one or both of these from occurring (Kung et al., 1998). Many additives have been developed to improve the ensiling process and the nutritive value of an ensiled forage (Leaver, 1990; Weinberg et al., 1993). Additives are expected to ensure a more efficient fermentation phase as well as reduce the risk of aerobic deterioration (Pettersson, 1988; Jonsson, 1989; Lättemäe, 1997). Therefore, silage additives used to improve silage quality should not only promote a rapid acidification via lactic acid, but also prevent the growth of spoilage organisms (Oude Elferink et al., 1999).


CHEMICAL PRESERVATIVES

Various chemical preservatives have been used to prevent spoilage when silages are exposed to air (Woolford, 1975; Kung et al., 1998). Significant improvements in nutritive value, the fermentation process, and aerobic stability can be achieved by the use of chemical additives such as organic and inorganic acids and nonprotein nitrogen (Steen, 1990).

Formic, acetic, propionic and butyric acids and higher volatile fatty acids as valeric and caproic can increase the stability of silage (Ohyama and McDonald, 1975; Ohyama et al., 1975; Woolford, 1975a; Woolford, 1978; Ashbell and Lisker, 1988; Detmer et al., 1999). Acetic acid in combination with lactic acid has been found to inhibit the growth of yeasts (Moon, 1983).

Henderson et al. (1979) reported that silages tended to be more stable when the total acid content was high.

The ability of organic acids to inhibit mold and bacterial growth is species and acid specific. It does not depend on pH, but rather the specific acid anion. For example, it has been shown that sorbic acid is quite effective in preventing the growth of Monascus spp., while acetic acid is totally ineffective against Monascus spp. (Schlatter and Smith, 1999). Furthermore, the effect of organic additives in reducing the susceptibility of a silage to aerobic deterioration has been found to vary (McGechan, 1990).

The inhibitory effects of propionic, butyric, acetic and sorbic acids on yeast growth and the effects of various levels of propionic acid in preventing silages from deterioration have been investigated (Ohyama et al., 1975).

Of the short chain fatty acids, propionic acid has the greatest antimycotic activity (Woolford, 1975a; Kung et al., 1998). Decreased deterioration with propionic acid has been observed in some cases (Kung et al., 2000).

However, Ohyama et al. (1975), observed cases where the presence of large amounts of propionic acid failed to prevent silage deterioration. No explanation was given for this contradictory finding; but they mentioned that it did not depend on DM content, total acids, lactic acid, WSC or pH.

Ohyama et al. (1975) reported that silages containing certain amounts of butyric acid or higher VFA did not exhibit deterioration. Because butyric acid production is an indication of poor quality silage, it is not recommended that silages of high butyric acid content should be made. It has been reported that the addition of isovaleric acid or caproic acid can prevent or delay aerobic deterioration (Ohyama and McDonald, 1975).

The use of formic acid results in unstable silages (Ohyama and McDonald, 1975). The application of formic acid or formaldehyde leads to high WSC remaining in silages. Di Menna et al. (1981) reported that formaldehydetreated silage after exposure to air underwent heating accompanied by high counts of aerobic organisms, both bacteria and yeasts. The formaldehydetreated silage was found to be more prone to deterioration than the untreated silage. The poorer aerobic stability of the treated silage was caused by the higher pH and greater WSC content that supported rapid microbial growth when exposed to air.

Henderson et al. (1972) observed an increase in the growth of yeasts in formic acid treated silages. Barry et al. (1980) reported that the onset of a rise in temperature in silages made with formaldehyde-containing additives was initially caused by bacteria (probably acid-tolerant) and that molds appeared secondarily.

Moderate levels of ammonia have improved the aerobic stability of corn silage (Kung et al., 2000), but Bolsen et al. (1992) summarized data showing that treating forages with ammonia decreased the recovery of DM resulting in poorer animal performance compared with untreated silages. Other research indicates that aerobic stability of grass silages increases when hexamethylene-tetraamine (HMTA), sodium nitrite (NaN), sodium benzoate and sodium propionate are used alone or together. At higher pH NaN is not very effective against spore-bearing bacteria, but HMTA is effective. Sodium benzoate is effective against mold even at pH 5 (Woolford, 1975b; Lingvall and Lättemäe, 1997).


BACTERIAL INOCULANTS

At present biological additives are preferred because they are non-toxic, non-corrosive to machinery, do not present environmental hazards and are regarded as natural products (Weinberg et al., 1993).

Inoculation with homofermentative lactic acid bacteria has improved silage quality. However, aerobic stability has often been worse with homolactic fermentations. The main advantage of homofermentative strains is their ability to promote fast and efficient lactic acid production, which results in a rapid decrease in pH. This does not necessarily relate to improved aerobic stability at feedout. The shift in fermentation products should not only improve silage quality (Kung et al., 1999), but also improve DM recovery.

Homofermentative fermentation should result in very little loss of DM while on the contrary, DM loss from heterofermentation can be substancial (Muck, 1996).

Bacterial inoculants have produced variable results on aerobic deterioration (Harrison et al., 1999; Holzer et al., 1999). A possible explanation for the aerobic instability of inoculated silages is that under strict anaerobic conditions, homofermentative LAB inoculants produce only lactic acid, whereas the naturally-occurring, heterofermentative LAB produce various volatile fatty acids (VFAs) (e.g. acetic, propionic, butyric). These shortchain fatty acids possess antimycotic activity, and thus inhibit the growth of yeasts and molds when these silages are exposed to air (Moon, 1983). The presence of low concentrations of oxygen in the silage results in a shift from homolactic to heterolactic fermentation (McDonald et al., 1991 as cited by Weinberg et al., 1993), which leads to the production of the VFAs with antimycotic activity. Mayrhuber et al. (1999) reported the highest aerobic stability in a silage with the highest concentration of acetic acid (up to 5.1% of the DM).

A high concentration of lactic acid does not always have a positive effect on aerobic stability. Yeasts are known to grow at low pH (Lindgren et al., 1985; Ruppel, 1997). Some yeasts can tolerate values below pH 3. The inhibition appears to be a function of lactic acid concentration. Woolford (1975b) determined a threshold value of 250 mMol lactic acid for inhibition of yeast growth. In Lactobacillus-treated silages, lactic acid concentrations reached 400 mMol. This effect may explain the decrease in yeasts in all inoculated silages after seven days of fermentation when lactic acid concentration reached its highest level. After that, selection led to a predominance of acidophilic yeasts. Also, there is no doubt that in treated silages, inoculated bacteria and yeasts compete for sugars. (Kibe et al., 1977). Cai et al. (1999) reported that the number of total yeasts was high in the LAB-treated silages, most strains had a high tolerance to lactic acid, and most were able to assimilate lactic acid and WSC during exposure to air, all of which lead to aerobically instability.

Recently, bacterial inoculants containing Lactobacillus buchneri effectively inhibited the growth and activity of spoilage yeasts and improved the aerobic stability of corn silage (Ranjit et al., 1998; Driehuis et al., 1999; Kung et al., 1999). The inhibition of spoilage yeasts is likely due to the ability of L. buchneri to ferment lactic acid to acetic acid and 1,2- propanediol. Recent studies suggest that L. buchneri can also produce other yet unidentified metabolites with antifungal activity (Oude Elferink et al., 1999). The efficacy of L. buchneri has been greater under laboratory scale than farm-scale conditions; and very few studies to date have measured the effect of L. buchneri on DM recovery, feed intake, and livestock performance.


ENZYMES

Enzymes such as cellulases, hemicellulases, xylanases, pectinases and amylase have also been used as silage additives. In theory, they break down complex polysaccharides (e.g., fiber and starch) to fermentable sugars causing the silage to be more digestible and providing more substrate for LAB to use in producing lactic acid (Kung and Muck, 1997; Muck and Kung, 1997). Enzymes have been more beneficial with legumes and grasses, which usually have a lower WSC content than corn, sorghum, or small grain cereals. Enzymes have usually been more effective in immature than in mature forages, which is likely explained by the greater lignification of cell walls in mature plants (Van Vuuren et al., 1989; Nadeau et al., 2000).

The potential benefits of enzymes include improved silage quality as measured by a reduction in the cell wall components, increased residual WSC, and higher DM digestibility. Improved silage quality should lead to improved livestock performance, e.g., increased feed intake, improved feed efficiency, faster daily gain, and increased milk and milk solids production (Stokes, 1992; Stokes and Chen, 1994; Hoffman et al., 1995; Kung and Muck, 1997).

A possible negative impact with enzymes could be that the extra sugars would support the growth of aerobic microorganisms, which could lead to a higher loss of nutrients during the feedout phase in aerobically unstable silages. In a review by Muck and Kung (1997), the authors found that the effect of enzymes on aerobic stability was variable and inconsistent over a wide range of forages and ensiling conditions.


Conclusions

Present evidence gives little information on the potential of a specific forage to ensile well or on the quality and stability of the end product (Woolford, 1984; 1990). Research into the processes of aerobic deterioration of silage has not explained why silages differ in their susceptibility to aerobic deterioration during the feedout phase (Barry et al., 1980).

It has been difficult to find silage traits that specifically predict aerobic instability or stability. Research has assessed how physical properties of silage affect aerobic deterioration; and although changes in microbial population by a restricted range of yeasts and bacteria have been observed, there is not a complete understanding of the microbiology of aerobic deterioration (Rees et al., 1983; Woolford, 1984; 1990; Pessi and Nousiainen, 1999).

Temperature, DM and WSC content, microbial population, and concentration of fermentation products in interaction with pH have the greatest direct effects on aerobic stability (Woolford, 1990; Pitt et al., 1991).

As mentioned previously, generalization about the microbial dynamics of aerobic deterioration is not possible. The microbial dynamics probably depend on the establishment and/or survival of the aerobic spoilers during the fermentation and storage phases, but at present little is known about the factors that influence this establishment (Woolford, 1984; Lindgren et al., 1988). The identity of the group of microorganisms that are responsible for the initiation and subsequent extent of aerobic deterioration can be affected by differences in the traits of a silage. Silages differ in their susceptibility to the loss of nutrients during the feedout phase, depending on how they were managed (Ruppel, 1997).

Potential for future developments to decrease problems associated with microbial degradation of silage lies in improved storage techniques and new additives (Lindgren et al., 1985; Clarke, 1988).