MAINSTREAM CIDERMAKING






1. HISTORY AND DEFINITION

Cider is generally regarded as a drink made from apples. The greatest production of fermented cider is in England (ca. 500 M litres p.a.), stretching in a band from the West Midlands counties of Hereford and Worcester through Gloucestershire to Somerset and Devon. Individual local operations are also found in East Anglia, Kent and Sussex. Production in France is restricted to the northwestern areas of Normandy and Brittany and is ca. 125 M litres p.a.. Production in Germany is centred on a Trier/Frankfurt axis and is of much the same volume. Smaller operations are found in Spain, Ireland, Belgium, Austria and Switzerland. In North America the term 'cider' refers to unfermented apple juice, but very limited quantities of 'hard' (i.e. fermented) cider are sold in upstate New York, in Quebec and in British Columbia. It is worth noting, however, that ca 125 M litres of fermented cider per year are produced in North America for direct conversion to vinegar (Lea 1989) In England, 90% of cider production is now concentrated in the hands of three large manufacturers. In France, one company presently owns two-thirds of the business through a large number of traditional brand names.

English cider is generally sold bright, in bottle or can, as an artificially sweetened and carbonated product with an alcohol content between 1.2% - 8.5% v/v. These are the UK legal limits (regulated in detail by Customs and Excise Notice 162). Brand names are particularly important since each company sells a wide range of products. Ciders are also available on draught for pub consumption. Some smaller new 'traditional' manufacturers also offer high-quality still or lightly carbonated ciders with heavier and more complex flavour characteristics than the large producers - often these are based on a defined blend of known cider apples which forms a positive selling point.

The diversity of UK cider styles reflects the relatively broad (and voluntary) British definition as "a beverage obtained by the partial or complete fermentation of the juice of apples....or concentrated apple juice....with or without the addition before or after fermentation of sugar or potable water" (NACM 1992). In France and Germany the definitions are more restricted by legislation, although the products in those two countries are very different (Possman and Lea 1992). French ciders are characterised by sweeter, tannic and apple-like flavours, whereas German ciders are relatively dry and acidic to an English palate.

Cider is in effect an apple wine, and good practice in wine and cider industries is similar. The world literature on cider-making is scant indeed. Most of it originates from the Long Ashton Research Station (LARS), near Bristol, which opened in 1903 as a cider research institute and closed that part of its work in 1986 due to cutbacks in government funding. However, the present success of the UK industry is largely attributable to the underpinning research conducted at LARS during those years, and much of the knowledge acquired over that time was authoritatively reviewed by Beech and Davenport (1970), Beech (1972), Beech and Carr (1977) and more recently by Beech (1993) and Lea (1995).
 

2. RAW MATERIALS

Apples are the primary raw material for cidermaking and the traditional classification for English cider apples is shown in Table 1.

Table 1- Classification of Cider Apples
Classification
Acid (%)
Tannin (%)
Sharp
> 0.45
< 0.2
Bittersharp
> 0.45
> 0.2
Bittersweet
< 0.45
> 0.2
Sweet
< 0.45
< 0.2

'True' cider cultivars, because they are selected solely for this purpose, have a number of advantages to the cidermaker. Chief amongst these are:

A distinguishing feature of true cider fruit, particularly French and English bittersweets, is the relatively high concentration of polyphenols, loosely known as 'tannin'. Although modern ciders are generally lower in tannin than in the past, it still makes an important contribution to overall mouthfeel of the beverage and prevents it becoming too insipid. The polyphenols make a major contribution to flavour, colour, pressability and also have weak anti-microbial properties (Lea 1984).

In some cases, cider fruit is also characterised by 'vintage quality', which is of particular concern to the small traditional producer. Vintage quality fruit gives generally more complex and interesting flavours to the cider than does bulk fruit. However, the vintage cultivars have generally lower yields and are more difficult to grow. Typical cultivars (bulk and vintage) are given in Table 2. Further descriptions of these and their cultivation are given in Williams and Child (1965), Williams (1987) and Williams (1988).
 
 

Table 2 - Typical Cider Apple Cultivars
Character
Early Season
Mid / Late Season
Sharp/Bittersharp Breakwells Seedling Brown's Apple
  Backwell Red* Frederick*
    Crimson King*
    Kingston Black*
  Foxwhelp Stoke Red*
     
Bittersweet Ashton Bitter Dabinett*
  Ellis Bitter Chisel Jersey
  Major* Harry Masters Jersey*
  Tremlett's Bitter Yarlington Mill*
  Taylors Michelin
    Vilberie
    Medaille d'Or*
     
Sweet   Sweet Coppin*
    Sweet Alford*
    Northwood*

* - Denotes vintage quality cultivars

It is rare for cider to be made of a single cultivar apple only. This is partly because the balance of sugar, acid and tannin required for a successful product is difficult to achieve from any single cultivar, and so a blend to achieve the appropriate balance is nearly always necessary. In addition, orcharding considerations such as the need for cross-pollination and a spread of harvesting period dictate the growth of relatively mixed orchards. Most large cider companies maintain a mixture of orchards under their own direct control as well as contracts with outside growers. The newer 'traditional' cidermakers, though they may start with fruit from existing orchards, are tending where possible to establish orchards of preferred cultivars under their own control.

Apple juice concentrate (AJC) is now widely used in UK cidermaking and is permissible also to a limited extent in France. Some UK companies work almost entirely from this source, but most use a mix of fresh juice and re-diluted AJC as required. The proportion of true cider fruit is sometimes very low, however, and many modern English ciders have a high proportion of dessert and culinary outgrades (particularly Bramley or Cox). Fermentable sugars from cane, beet or hydrolysed starch are also extensively used as adjuncts in modern UK cidermaking.
 

2.1 Juice Preparation

Fresh fruit should be fully ripe before cidermaking and is generally stored for a few weeks after harvest so that all the starch can be be converted into sugar. Apples must be sorted and washed before milling to eliminate rotten fruit and orchard debris which have adverse effects on microbiological status and ultimate cider quality. Pack presses have traditionally been used for juice preparation but horizontal piston presses and/or belt presses are now standard in the larger factories. Once the juice is prepared, it is coarsely screened and run off to tanks of fibreglass, high density polythene, stainless steel or (less commonly) wood for pre-fermentation blending and additions.

Before fermentation the juice must be prepared accordingly. The traditional spontaneous clarification of juice by 'keeving' ('maceration et cuvage' in France) is now obsolete in England (Beech 1993, Lea 1995). In modern English cidermaking the fermentable sugar sources (juice, AJC and syrups) are blended to the required level. This may be as high as S.G. 1.080 - 1.120 to give a final alcohol of 10 - 15 % (which is then diluted before retail sale). Nutrients are also added to ensure a complete and speedy fermentation to dryness, unlike the traditional keeving where nutrients are removed from the juice to ensure a slow fermentation. It is usual to add ca 250 ppm ammonium sulphate or phosphate, and thiamin at 0.2 ppm. Pantothenate, pyridoxine and biotin may be useful too. These additions are particularly important if the juice is made up with fermentable adjuncts (which do not contain any nutrients) or with AJC, since the Maillard reaction during concentrate storage reduces nutrient levels and generates yeast inhibitors such as 5-hydroxymethylfurfural (HMF).

If clarified concentrates and adjuncts are to be fermented, a source of insoluble solids is often helpful. This allows the yeast cells a solid surface on which to rest, and from which ethanol and CO2 can be liberated to the medium. This may be achieved by the addition of bentonite at say 0.5% to the must before fermentation, which also aids the subsequent clarification of the cider (Ough and Groat 1978). Addition of sterol precursors may also be useful to encourage healthy development of yeast cell walls, although this may also be promoted by yeast aeration after pitching (Ewart 1995).

Many cidermakers routinely add a pectolytic enzyme preparation prior to fermentation of fresh juice, to prevent post-fermentation hazes (AJC is of course already depectinised during manufacture). Pectolytic enzymes are sometimes added initially to the fruit pulp, if cull apples such as Cox are in use, to enhance pressability and to increase yield as well (Lea 1991, 1994).

The most significant adjunct in modern UK cider making, as in white wine making, is sulphur dioxide. The effectiveness of SO2 is pH dependant since it is only the undissociated form (so-called 'molecular SO2') which has anti-microbial properties. Hence cider juices should always be brought below pH 3.8 by the addition of malic acid before SO2 addition and the amount to add should be reckoned from Table 3. With healthy fruit containing only small amounts of sulphite-binding components, this should leave sufficient free SO2 to provide an effective sterilisation before the addition of a yeast inoculum 12 hours later. If the original fruit is in poor condition, it may contain large amounts of 5-ketofructose or diketogluconic acid from bacterial activity which will bind most of the added SO2 and reduce its effectiveness (Burroughs and Sparks 1964, 1973). Oxidised ascorbic acid, native to the apple, will also degrade to L-xylosone which is a strong sulphite binder too. Recent unpublished work has shown that modern juices made up from depectinised apple juice concentrate contain relatively large amounts of free galacturonic acid. Although this is only a weak sulphite binder its effect becomes significant at the high concentrations (thousands of ppm) which are present.
 

Table 3 - SO2 addition required to cider apple juices
pH
Addition required (mg/l)
3.0 - 3.3
75
3.3 - 3.5
100
3.5 - 3.8
150

Juices of pH > 3.8 (as in many full bittersweets) should be brought
down to this value by blending or acid addition and 150 ppm SO2 then added.
 

3. FERMENTATION
 

3.1 Yeast Selection

In traditional cidermaking no external source of yeast is added. However, since the apples themselves contain a mixed yeast microflora which may be in the order of 5 x 104 cells per gram of stored fruit, spontaneous fermentation will commence within a few hours if the temperature of the juice is above 10o C (Beech 1993). In a traditional cider fermentation, where no yeast is added and no sulphite is used, the first few days are dominated by the non-Saccharomyces species such as Kloeckera apiculata and Metschnikowia pulcherrima. These multiply quickly to produce a rapid evolution of gas and alcohol. They also generate a distinctive range of flavours, characterised by ethyl acetate, butyrate and related esters. As the alcohol level rises (2 - 4%), these initial fermenters begin to die out and the microbial succession is taken over by Saccharomyces uvarum. If sulphur dioxide is added to the initial juice, the non-Saccharomyces yeasts and most bacteria are suppressed or killed. This allows the Saccharomyces species to multiply after a lag phase of several days, and the fermentation then proceeds to dryness with a more homogeneous and benign microflora than in the case of an unsulphited juice. Secondary infection by film yeasts and acetifying bacteria is also less likely.

Nowadays, however, few cidermakers in the UK wait for the naturally selected Saccharomyces to establish themselves. Since the 1980's the use of active dried wine yeast has become almost universal in the mainstream UK cider industry. A mixed inoculum of S. uvarum and bayanus is often used, on the grounds that the first yeast provides a speedy start but the second will cope better with the fermentation to dryness of the high alcohol bases which are now common throughout the industry. These dried yeast require no pre-propagation and are simply hydrated in warm water before pitching directly into the juice.

Traditional cidermakers, or those who are hoping to re-establish tradition, do not necessarily follow suit on yeast inoculation and may prefer some element of the natural microflora to remain. In Germany there has been some concern that fermentations dominated entirely by Saccharomyces are lacking in estery cider character (the so-called 'Apiculatus-ton'), and that the role of Kloeckera apiculata is important (Schanderl et al 1981, Scholten 1992). Similarly, in France the need for a mixed microflora is regarded as axiomatic and recent experimentation has focussed on mixed inocula of e.g. Metschnikowia pulcherrima and S. uvarum in an attempt to produce a complex and traditional flavour but under closer microbiological control (Bizeau et al 1992). Modern French factory cidermaking is still based on traditional procedures, and care is taken to ensure a cool slow fermentation so that significant residual sugar remains in the final cider (Revier 1985, Drilleau 1988, 1989).

English factory practice is almost completely the opposite. The juices are prepared and inoculated as described above, and then a rapid and complete fermentation to absolute dryness is encouraged. Although in most cases there is no formal temperature control, a range of 15o- 20o C is considered desirable. Thus a portion of the fermenting juice is sometimes warmed to 25o C by pumping through an external heat exchanger if it is slow to start or to finish. Most large UK cidermakers have recently taken the view that a complete fermentation to 10 - 12 % alcohol in as little as two weeks would be a desirable objective. However, there are signs that this attitude may be changing, since the flavour quality and stability of the finished ciders can be compromised under such stringent conditions.
 

3.2 The malo-lactic fermentation

Traditional ciders are very frequently subject to a malo-lactic fermentation. As in wines, the major desirable organism effecting this change appears to be the heterofermentative coccus Leuconostoc oenos, although other Lactobacillus species may also be present (Beech and Carr 1977, Carr 1983, 1987, Salih et al 1988). It is favoured by a lack of sulphiting during fermentation and storage, and by a certain amount of nutrient release from yeast autolysis when the cider stands unracked on its lees. In French cider making where the primary fermentation is very slow, the malo-lactic change may occur concurrently with the yeast fermentation, whereas in UK cider-making it is most likely to occur once the yeast fermentation has finished and the cider is in bulk store (Drilleau 1992).

The most obvious external feature of the malo-lactic change is the decarboxylation of malic to lactic acid and the consequent evolution of gas. The acidity also falls and the flavour becomes rounded more complex. In modern UK factory cidermaking, the malo-lactic fermentation is generally regarded as a nuisance and is not encouraged. In any case the prevailing conditions do not favour it, since sulphite is generally used before and after fermentation and the ciders do not stand on their yeast lees for long. Nonetheless, spoilage of stored ciders by rod-shaped lactic acid bacteria is not uncommon and often manifests itself nowadays by blockage of membrane filters during final packaging. However, the organisms involved in this case are not usually those which are associated with the traditionally desirable effect of the malo-lactic change.
 

3.3 Sulphite binding

The binding of added sulphur dioxide to juice carbonyls has already been described, but the principal source of sulphite binders in cider is generated during fermentation by the normal process of glycolysis and the operation of the Krebs cycle (Whiting 1976). Pyruvate, a-ketoglutarate and acetaldehyde are all essential metabolic intermediates in the production of ethanol by yeast. However, they are all carbonyls which bind to sulphur dioxide, and the amounts remaining at the end of fermentation will impact directly on the efficiency of any sulphite which is added to the cider for storage (Burroughs and Sparks 1964, 1973). Acetaldehyde is by far the strongest binder and until all this component is bound no free sulphite can in practice remain in the cider. The other carbonyls bind less strongly and hence can co-exist partly unbound in equilibrium with free SO2. The percentage of cider carbonyls which are bound at a typical level of 50 mg/l free SO2 is given in Table 4.
 

Table 4 - Binding of SO2 to juice carbonyls
Compound Percentage of carbonyl that is bound Typical level in cider (mg/l) Bound SO2 contribution (mg/l) due to that carbonyl
Naturally present      
Glucose 0.11 7000 8
Galacturonic acid 4.4 1000 15
L-xylosone 36 20 4
Acetaldehyde 99.8 25 35
Pyruvate 83 20 12
a-keto glutarate 58 15 4
       
From bacterial contamination      
5-keto fructose 70  should not   
2,5 diketo gluconic acid 64  be present  
       
OVERALL BOUND SO2     78
TOTAL SO2 (bound + free)     128

The carbonyl-bound sulphite has little anti-microbial action and yet it is determined as part of the total sulphur dioxide when legislative limits are to be complied with. Reduction in the total amount of SO2 can only be achieved by minimising the amounts of sulphite-binding carbonyls.

It is known that the addition of thiamin, for instance, will reduce the production of pyruvate and a-ketoglutarate during fermentation, since thiamin is an essential co-factor in the conversion of pyruvate to ethanol. It is also known that acetaldehyde production is reduced by added pantothenate. Different yeast strains produce inherently different levels of the main sulphite binders too, and it is an active topic of research to establish how these strains and their interaction with fermentation conditions can be manipulated to produce the minimum amount of carbonyls. It is also known that the malo-lactic fermentation can help to reduce sulphite binding capacity due to loss of pyruvate.
 

3.4 Cider Colour

The colour of cider is determined by juice oxidation or degradation and in fact it is possible to make water-white high tannin ciders if oxidation is completely inhibited (Lea and Timberlake 1978, Lea 1982). The effect of pulp and/or juice oxidation on juice colour sets the primary appearance of the juice which is due to the quinoidal oxidation products of phloridzin, epicatechin and the procyanidins (Goodenough and Lea 1979, Goodenough et al 1983, Lea 1984, Lea 1991). During yeast fermentation, however, the initial colour of fresh juice diminishes by around 50%, due to the strong reductive power of yeasts, which readily convert quinone groups to hydroxyls with consequent loss of the chromophore. The colour from concentrate drops only 10% or so during fermentation, however, since the carbonyl-amino chromophores from Maillard browning are resistant to this reducing action.
 

3.5 Cider Flavour

As with any beverage, the flavour of cider is a combination of taste and aroma. Traditional English and French ciders made from bittersweet fruit have been distinguished by relatively high levels of bitterness and astringency due to the polyphenolic procyanidins (tannin). The oligomeric procyanidins (n = 2-4) are more bitter ('hard tannin') than the polymeric procyanidins (n = 5-7) which are the more astringent ('soft tannin'). The levels are initially set by cultivar - thus 'Tremletts Bitter' is more bitter than the astringent 'Vilberie' although both fruits have the same level of procyanidins in total (Lea and Arnold 1978, 1983). Juice processing conditions (notably oxidation) also play a part in determining the final non-volatile flavour, since oxidising procyanidins become 'tanned' onto the apple pulp and both bitterness and astringency can markedly diminish (Lea 1984, 1990). Nowadays the heavily tannic flavours of traditional ciders are much less in demand and the procyanidins are noticeable in modern factory ciders only as a part of the general mouthfeel.

The volatile flavour of cider is in most part qualitatively identical to that of all other fermented beverages and derives to a large extent from the yeast utilising well-known biosynthetic pathways (Berry 1995). Ciders have traditionally been regarded as high in 'fusel alcohols', particularly 2-phenyl ethanol, which has often been attributed to their low nutrient status. It is also known that higher fusel levels are generated from cloudy rather than clear juice fermentations (Beech 1993).

Detailed work at LARS over a number of years listed several hundred compounds as contributors to cider flavour (Williams and Tucknott 1971, 1978, Williams 1974, Williams et al 1978, 1980, Williams and May 1981). Some of these may arise from non-volatile glycosidic precursors which are hydrolysed by enzymic action when the fruit is disrupted. Therefore the high levels of 2-phenylethanol and its esters in ciders may not derive de novo from yeast synthesis (although this route is known), but from the presence of a glycosidically bound form in the fruit which is liberated and cleaved during fermentation (Schwab and Schreier 1988, 1990).

One of the most interesting and perhaps unique volatile components of cider was described by Williams et al (1987) and also by Hubert et al (1990). Unpublished work in our own laboratory using 'odour-port dilution analysis' has shown that it has the lowest threshold and therefore the greatest single odour contribution of any cider volatile. It also has a distinctive 'cidery' aroma. Its molecular mass is 172, for which a number of structures have been proposed including the acetal 1-ethoxyoct-5-en-1-ol (Williams et al 1987). In our view it may instead be the dioxolane which results from the condensation of acetaldehyde with octane-1,3,-diol. The diol itself is a relatively unusual alcohol which is known to be present in apples in a glycosidically bound form and can reach levels of 100 ppm in stored fruit (Berger et al 1988). Such a 'cidery' component therefore appears to result specifically from the action of alcoholic fermentation on apples.

A further group of components results from the malo-lactic fermentation. It is well known that diacetyl is synthesised from pyruvate by Leuconostoc species and contributes positively to 'buttery' flavours in wines and ciders. Another group of flavours described as 'spicy' and 'phenolic' derives principally from the malo-lactic fermentation in bittersweet ciders. These are typified by ethyl phenol and ethyl catechol which arise from hydrolysis, decarboxylation and reduction of p-coumaroyl quinic and chlorogenic acids respectively (Beech and Carr 1977). Although these volatile phenols are not unique to cider, being found in whiskies too, they are distinctive contributors at low levels to the characteristic 'bittersweet' flavours of well-made traditional ciders from the West Country or Normandy.
 

4. POST FERMENTATION OPERATIONS
 

4.1 Racking and Storage

Once fermentation is complete, ciders are racked from the yeast lees for storage. Current practices vary widely. In some English factories, racking and clarification takes place as soon as possible for virtually immediate blending and packaging without any maturation. In others, the ciders remain on their lees for several weeks and are racked into inert tanks or oak vats for a maturation period of several months. During this time a malo-lactic fermentation may or may not be encouraged - if considered desirable, no sulphur dioxide must be added during storage.

Initial clarification may be performed by the natural settling of a well-flocculating yeast, by centrifugation, by fining or by a combination of all three. Typical fining agents are bentonite, gelatin, isinglass or chitosan (a partially de-acetylated chitin prepared from crab-shell waste in North America or the Far East). Gelatin is often used together with bentonite to encourage floc formation. The use of the highly efficient gelatin/kieselsol system is widespread in Germany but less common in the UK, where instances of 'over-fining' and persistent gelatin hazes therefore sometimes occur. Ciders made from cloudy juice concentrates can often prove extremely intractable to fine and may give persistent hazes.

Nearly all UK ciders are blended before sale, although small 'traditional' cidermakers generally perform less blending and manipulation operations at bottling than do the major producers. In a large factory, there may be dozens of different fermentations running or maturing concurrently, from different juice sources and intended for different products. These form the high alcohol 'base ciders' from which blending is performed according to the cidermaker's requirements. Water will be added to these bases to give the correct alcoholic strength for retail sale, together with additions of sugar and artificial sweeteners, malic or other acids, permitted food colours, preservatives and carbonation. Generally, UK regulations permit for cider all those operations or additives which are allowed by EU 'horizontal' food law. In France and Germany, specific 'vertical' legislation applies to cider and is much more restrictive.

Final filtration may take place just before and after blending. Generally, powder filters or coarse disposable sheets are used to produce a bright product, followed by near-sterile sheet or membrane filtration (0.5 mm) to remove all yeasts amd most bacteria. Most ciders are then pasteurised and/or carbonated into the final pack. In some cases, in-bottle or tunnel pasteurisation of glass bottles or cans is still used. In other cases, the cider is hot filled into glass. With the increasing use of PET bottles in most large factories, HTST treatment in a flow-through pasteuriser and chiller is required followed by near-aseptic filling conditions. Alternatively, cold aseptic filling after sterile membrane filtration (0.2 mm) is used. Nearly all cider makers will aim to add 50 ppm of SO2 at filling to give an equilibrium level of 30 ppm free SO2 in the beverage. This depends on the level of sulphite binders in the cider as described earlier. For cans, the total level of SO2 compatible with the lacquer is often as little as 25 ppm, and so ciders destined for canning are often specially fermented in the absence of sulphite throughout.

There is a certain market for 'naturally conditioned' ciders in kegs or small plastic barrels. These are generally produced from fully fermented dry ciders, to which an additional charge of sugar and flocculating yeast has then been added. The product is of course somewhat cloudy but may remain in good condition for many weeks due to the slow continued fermentation. True 'champagne' ciders, prepared by fermentation in bottle followed by 'disgorgement' of yeast from the neck, are virtually unknown now in the UK although some traditional cidermakers are attempting to revive the style.
 

4.2 Storage Disorders

The classical microbiological disorder of stored bulk ciders is known as 'cider sickness' or 'framboisé' in French (Beech and Carr 1977, Carr 1987). This is due to the sulphite-resistant bacterium Zymomonas anaerobia which ferments sugar in bulk sweet ciders stored at pH > 3.7. The features of cider sickness are a renewed and 'almost explosive' fermentation, accompanied by a 'raspberry' or 'banana-skin' aroma and a dense white turbidity in the beverage. It is virtually unknown in English cidermaking today since the ciders are generally at pH < 3.5 and are never stored sweet.

Another classical disorder, termed 'ropiness', is still encountered. This is caused by certain strains of lactic acid bacteria (Lactobacillus and Leuconostoc spp.) which synthesise a polymeric glucan. At low levels this increases the viscosity of the cider and when poured it appears 'oily' in texture with a detectable sheen. At higher concentrations of glucan, the texture thickens so that the cider moves as a slimy 'rope' when poured from a bottle. The flavour is not much affected. If not too severe, ropy cider can be cured by agitating vigorously to break up the glucan chains, followed by the addition of 100 ppm SO2 to prevent further growth.

'Sub-acute' ropiness is relatively common and is a frequent though largely unrecognised cause of membrane filter blockage since (unlike depth filters) they have relatively little tolerance to the presence of small quantities of 'coating' polysaccharides. We have identified a number of such cases in recent years where the blocking agent, once isolated and characterised, proved to be of this type. In some cases the bacteria (both rods and cocci) could also be identified upstream of the membrane pre-filter, and the problem was traced back to inadequate sulphiting which caused the bacteria to proliferate in storage. Other related cases of filter blockage are attributable to mannans (which may derive from extracellular yeast polymers) or arabinans (from insufficiently degraded pectin side chains in apple juice concentrate). Fructans of unknown origin have also been identified in our laboratory as a recent cause of membrane filter blockage.

A sulphite-resistant spoilage yeast which is often poorly recognised grows slowly in sweetened bottled ciders to form large flaky clumps which do not necessarily appear to be yeast-like on initial examination. This is Saccharomycodes ludwigii, which originates from the cider fruit and displays particularly large cells (25 microns diameter). In bottle it is often mistaken for a so-called 'protein' deposit.

True protein deposits in bottled ciders are actually very rare because the native protein content of apple juice is so low (ca 100 ppm). Such deposits nearly always result from 'over-fining' at some point in production where excess gelatin has been added. Many apple juices which are used to prepare concentrate are fined with gelatin and kieselsol prior to concentration in their country of origin, so the concentrates when purchased may contain relatively large amounts of unstable protein still in the presence of colloidal silica. This will not be apparent in the concentrate itself since the protective effect of the high solids prevents agglomeration and flocculation occuring. After fermentation and dilution, however, the unstable protein may eventually precipitate to form a haze or a deposit.

Many cider deposits also involve significant quantities of polyphenols, in conjunction with protein and polysaccharide. This complexation may 'shock out' a haze or a deposit when the product is cooled. Even in the absence of protein, such hazes may still form at low pH due to the breaking and re-forming of carbon-carbon bonds between the procyanidin units, leading to the slow build-up of random polymers which eventually drop out of solution.

Flavour taints in ciders may arise from adventitious contamination e.g. the presence of naphthalene where tarred rope had been stored adjacent to a cider keg. Many taints are however endogenous or arise from an imbalance in the natural flavour profile due to microbiological action. For instance, volatile alkyl phenols are normal and desirable constituents of ciders at low levels but can become overt taints if unwanted bacterial action generates large amounts from their non-volatile precursors. A wide range of 'sulphidic' and 'woody' notes are associated with modern commercial ciders and sometimes become regarded as taints. However, they appear to have extremely low thresholds (parts per trillion or less) and their nature remains unknown.

A frequent cider taint is that of 'mousiness'. This was extensively investigated by Tucknott at LARS (1977) and more recently by Heresztyn and colleagues in Australia (Strauss and Heresztyn 1984, Craig and Heresztyn 1984, Heresztyn 1986)). Isomers of 2-acetyl or ethyl tetrahydropyridine are the tainting species, generated probably by the growth of Lactobacillus or Brettanomyces spp. under aerobic conditions in the presence of both lysine and ethanol. As bases, they exist in the salt form in ciders and are not detectable until converted to the free base (volatile) form in the mouth. Hence mousiness is rarely detectable in the headspace aroma of ciders, and takes a few seconds to appear when the cider is tasted. Analysis by odour-port gas chromatography in our own laboratory has confirmed that several closely related compounds are in fact present in 'mousy' ciders.

A newly described taint in ciders is that due to indole. This compound is well-known in meat products, particularly pork, where it can form a part of the so-called 'boar taint' and is derived from tryptophan breakdown (Wilkins 1990). Work in our own laboratory has identified indole as a relatively widespread taint in ciders, which may derive from an odourless precursor or salt since it often appears and disappears from bottled products. Current belief is that it is generated de novo by the yeast from inorganic nitrogen during its own synthesis of tryptophan, rather than its breakdown. Its presence in English factory ciders is probably due to the unusual nutrient status of the fermenting juice compared to traditional practice.
 

5. REFERENCES
 

Beech F.W. and Davenport R.R. (1970) The role of yeasts in cider making. In The Yeasts (1st edn.) Vol.3. 'Yeast Technology'. Rose A.H. and Harrison J.S. (eds). Academic Press, London. pp 73 - 146.
 
 

Beech F.W. (1972 a) English Cidermaking - Technology, Microbiology and Biochemistry. In Progress in Industrial Microbiology. Hockenhull D.J.D. (ed). Churchill Livingstone, London. pp 133 - 213.
 
 

Beech F.W. (1972 b) Cider making and cider research - a review. J. Inst Brewing 78 477
 
 

Beech F.W. and Carr J.G. (1977) Cider and Perry. In Economic Microbiology Vol 1. 'Alcoholic Beverages'. Rose A.H. (ed). Academic Press, London. pp 139-313.
 
 

Beech F.W. (1993) Yeasts in Cider Making. In The Yeasts (2nd edn.) Vol 5. 'Yeast Technology'. Rose A.H. and Harrison J.S. (eds). Academic Press, London. pp 169 - 213.
 
 

Berger R.G., Dettweiler G.D. and Drawert F.(1988) Occurrence of C8 diols in apples and juices. Deutsch. Lebensm. Rundsch. 84 (11) 344 - 347.
 
 

Berry D.R. (1995) Alcoholic Beverage Fermentations. In Fermented Beverage Production Lea A.G.H. and Piggott J.R. (eds). Blackie, Glasgow. pp 32 - 44
 
 

Bizeau C., Le Quere J.M., Michel A. and Drilleau J-F (1992) Possibilities for modelling cider fermentation Industries Agricoles et Alimentaires109 7 - 9.
 
 

Burroughs L.F. and Sparks A.H. (1964) The identification of sulphur dioxide binding compounds in apple juices and ciders J. Sci. Food Agric15 176 - 185
 
 

Burroughs L.F. and Sparks A.H. (1973) The sul