Beer Stability and Head Retention

MEETING OF THE SCOTTISH SECTION, HELD AT THE NORTH BRITISH HOTEL, EDINBURGH, ON TUESDAY, 17th OCTOBER, 1950

Mr. A. Clark Doull in the Chair.
The following paper was read and discussed:

BEER STABILITY AND HEAD RETENTION
By E. Helm (Alfred Jorgensen Laboratory of Fermentology, Copenhagen, Denmark) Received 2nd November, 1950

On the basis of results of recent investigations of factors affecting beer stability, it is possible to make recommendations of procedures which will favour the production of stable beer. The troublesome β-globulin cannot be eliminated from barley during malting, but acidification of the mash to the isoelectric point of this protein (pH 4·9) gives encouraging results. Closed cooling systems favour stability, and avoidance of contact between the fermented beer and air is of the first importance. Kieselguhr is appropriate for pre-filtration, but centrifuging is not advisable when high stability is desired unless the problem of heating during treatment can be solved. Sheet filters alone should not be used for beers intended for pasteurization, but pulp filters are more efficient in removing chill-haze materials. Foaming properties are dependent on materials different from those responsible for non-biological stability, so that theoretically it should be possible to improve stability without impairing foaming, though at present the differential treatment necessary to achieve this cannot be realized to the full.

Head retention and stability of beer have been studied intensively during the last 10 or 15 years, and it might be of interest to review the present state of knowledge in this important field. It is, however, not the intention to deal with all the results of research in detail; the main purpose is to deal with those aspects which have relation to practice. A great deal of the work that has been carried out is fundamental research, and it is not yet certain what the consequences of these researches will be in brewing practice.

It is difficult to avoid mentioning a number of facts with which readers will be quite familiar. However, everyone will realize the necessity of mentioning them, since the new discoveries should be put into their proper places in the total picture.

STABILITY OF BEER
Beer is an unstable liquid which in time will necessarily become turbid. Turbidity may be due to the development of microorganisms or to the formation of non-biological sediments.

Biological Turbidities
The micro-organisms which develop in beer are yeasts or bacteria. All kinds of culture yeasts and wild yeasts are able to grow in beer, but with bacteria it is quite different, since of these only a comparatively small number of species are capable of developing. The reason for this is the acidity of beer, its contents of preservative hop constituents, and most likely the absence of certain growth factors. The most common beer-spoiling bacteria are rods (of which lactic acid rods are the most important), Pediococcus spp. and, especially in Britain, Acetobacter spp. Factors restricting development of yeasts are:
(a) hjgh attenuation of the beer (low finals);
(b) absence of air;
(c) high alcohol content.

As sugar is the main nutrient for yeast, the sugar contents should preferably be as low as possible. Under practical conditions as found today, especially in Britian, little can be done in this respect. In Continental brewing, and particularly in lager brewing, no primings are used, and the beer is fermented as near as possible to the limiting attenuation. This has proved of great importance to biological stability. About 1%, or even less, of fermentable sugar is left in such beers of an original gravity of 1,042 to 1,050, when they are being marketed without pasteurization. Absence of air in the finished beer is of no less importance, since development of yeast is greatly restricted if the oxygen content in the beer is low (de Clerck, 1934¹). Analytical control in this respect is most important, and air content in beer can be determined by Mendlik’s method.² This method is simple and quick and can be carried out in any brewery, and although the accuracy is not of the highest order, it is quite sufficient for the purpose.

In order to control the attenuation, and thereby the content of sugar, determination of the limiting attenuation is most important. This analysis is one of the most important analyses to be carried out in any brewery. Without knowing the limit or the limiting final to which a particular brew might be fermented, the brewer is working in the dark when conducting the fermentation.

Wild Yeast.— Recently some most interesting work has been published dealing with wild yeasts (e.g.. Wiles³). A number of wild yeasts have been isolated, and their proper Latin names have been determined. This work is important fundamental research. From a practical point of view there are some harmless wild yeasts and some harmful ones, as well as a number of intermediate forms. It would be most valuable if brewers were in a position to choose between the types of wild yeasts so as to select the harmless ones for their own breweries. Needless to say, this is not possible; nor is it possible to combat only the harmful wild yeasts, leaving the harmless yeasts and the culture yeast un touched. The only practical thing to do is to fight against all wild yeasts (and bacteria) at the same time, no matter whether the name be pastorianus or Pichia.

Continental brewers have been rather successful in their fight against wild yeasts, using pure pitching yeast in combination with proper rinsing and disinfection of the plant. Today wild yeast presents no serious problem at all in Continental breweries.

Bacteria. — The three restricting factors in connection with development of yeasts are also important with regard to the growth of bacteria. In particular a low sugar content and a high alcohol content will help in retarding bacterial growth.

A low oxygen content will influence bacterial development differently, for, as is well known, some bacteria are aerobic, and others are anaerobic. Among the former, Acetobacter is the most important. This bacterium presents no problem at all in lager brewing, and the common occurrence of Acetobacter in top fermented beer might be taken as an indication of high air content in British beers. However, Acetobacter infection should be rather easy to deal with, once the air content has come under control.

Acidity is an important restricting factor against bacterial development in beer. pH is as an average higher in lager than it is in British top-fermented beer (4·4-4·5 in lager as compared with 3·9-4·0 in ale). This fact might explain why bacterial infection in beer is more common on the Continent than it seems to be in Britain.

Preservative value of hop constituents is said to be of the greatest importance in British brewing, and the hops are selected according to P.V., whereas this hop quality is not regarded with the same respect on the Continent. In lager brewing there is at present an increasing number of breweries in which hops are selected according to bittering value, i.e., particularly α-acid content (humulone).

Growth substances such as bios, vitamins and the like are present in wort and play their important part in making beer wort such an ideal substrate for yeasts and for a number of beer-spoiling bacteria. After the fermentation and clarification of beer a certain number of these substances are lacking in the finished product. The yeast absorbs a number of them and excretes others. Under ideal conditions some of these substances are absent or at least present in but minimum concentration in the beer. There can be no doubt that such beer is not an ideal substrate for yeasts and bacteria, although there may be rather differing requirements.

On the other hand, if the separation of yeast from beer has been incomplete, and yeast autolysis has occurred, the beer will contain more growth substances both in number and concentration, making that particular beer a better substrate. Here is a point which might help to explain some of the irregularities which all brewers have experienced.

Non-Biological Turbidities
The main turbidities occurring in pasteurized beer are chill haze and oxidation haze (Helm, 1936⁴). Chill haze appears upon chilling of the beer in ice, disappearing again when the beer temperature is increased. This reversibility is characteristic of chill haze. Beer which has been kept beyond a certain time in bottle shows a permanent turbidity in addition to the chill haze. This turbidity is formed by oxidation of the chill-haze material, which is the reason for its name: oxidation haze.

There is no sharp distinction between the two kinds of sediment. Chill haze appears in all pasteurized beers (which have not been specially treated) within a few days of pasteurization, its intensity increasing with time and being accelerated by heat. Oxidation haze, however, does not appear in normal beers until about three weeks later.

The mother-substance of both hazes is the same, namely a compound formed by globulin particles with tannin, containing about one third tannin and two-thirds globulin. Thanks to the very important work carried out since 1945 at the Stockholm Breweries by Sandegren and coworkers, ^5,6,7 the natural history of haze formation in beer has to a large extent been made clear.

In 1942, Quensel⁶ had found that barley contains four different globulins which he called α, β, γ, and δ-globulin. As globulins, these are salt soluble, and with the exception of δ-globulin they are found in malt also.

β-Globulin is the one which is of interest for the present subject since α- and γ globulins are more or less eliminated in the course of the brewing process, and as far as is known they are unimportant with regard to haze formation in the finished beer. β-Globulin, however, remains nearly intact in the malt, and so it is not affected by the malting process. In the mashing process also, the β-gIobulin is almost unaffected by enzymes. This resistance towards enzymes during malting and mashing is the outstanding characteristic of β-globulin. Hence, the amount of β-globuhn in the wort is not influenced to any practically important extent by changes in the malting and mashing processes. This is in good agreement with practical experience.

In the wort, a certain amount of β-globulin is dissolved. As a result of the boiling, the β-globulin which has a molecular weight of 80,000-100,000 is split into smaller molecules, of probably one-half or one-third the size. These globulin fragments combine with tannin which is always present in the wort, and the compound thus formed appears as haze or turbidity when the wort is chilled. The main part of the fine or cool sludge is, according to Hartong,⁹ the same as the chill haze substance, and possesses its characteristic reversibility of behaviour with respect to dulling and heating. To some extent this chill-haze substance is separated during the processes of wort clarification, fermentation, storage and nitration. Without special means, however, it is not possible to eliminate this compound entirely; a certain amount will remain in the finished beer and, although it only represents a few ppm, is sufficient to make the beer liable to chill haze. When the pasteurized beer is stored, oxidation of the globulin-tannin compound takes place, and an insoluble sediment is formed which is called oxidation haze.

Sandegren^5,6 has shown that β-globulin (as distinct from the other barley globulins) as well as chill-haze substance is especially rich in sulphur (l·9-2·0 % of the protein content). Furthermore, high contents of active —SH groups are found, which on oxidation combine into —S.S— links. By this process a type of condensation takes place, and the larger oxidized molecules or particles, being insoluble, fall out of solution to form a sediment. In 1939 the present author¹⁰ isolated a considerable amount of oxidation-haze sediment from turbid beer. The protein part of this sediment was found to contain 2·8 % of sulphur. So sufficient evidence is at hand in support of the —SH, —S.S— theory which was first put forward by Hartong¹¹ in 1934.

The other part of the chill-haze substance, namely the tannin, is also known to be oxidizable. When tannin is oxidized, insoluble reddish-brown phlobaphenes are formed. The formation of oxidation haze is undoubtedly due to both these oxidation processes. Traces of heavy metals such as copper and iron have been found in the ash of chill haze substance and oxidation-haze sediment from beer. Some evidence is at hand (Helm¹⁰) which shows that these metals catalyse the formation of oxidation haze.

This, then, is the natural history of the two main turbidities occurring in pasteurized beer. This explanation is, of course, only an outline and is perhaps somewhat over simplified. Based on the picture just given, an attempt may now be made to consider what can be done in practical brewing in order to produce a stable beer.

Factors for Stability
Barley. — Barley is the only cereal which contains the unwanted β-globulin, as shown by Säverborn, Danielsson and Svedberg¹² Barley, however, is indispensable in brewing the types of beer with which we are concerned. The other cereals, e.g., rice and maize, that are in common use as raw materials in brewing contain less total nitrogen than malt, and it is now known that these raw grains do not contain the β-globulin; thus, it is confirmed that the prevailing practice of using such raw materials in order to improve the stability of the resulting beers is justified—they act simply as diluents for the unwanted β-globulin. At the same time, however, the other nitrogen compounds in the beer are diluted in the same proportion. The result of this, as is well known, is a lower head retention in beer brewed with adjuncts.

Recent results of Sandegren¹³ indicate that different barley varieties contain different quantities of β-globulin. His investigations were made by fractionation of barley extracts into different nitrogenous compounds followed by polarographic measurements of the fractions. If and when such determinations can be made more simply and easily there will be a valuable means of selecting barley with the lowest β-globulin content; the practical value of this is evident. Even more valuable would this method prove in malting barley research work. This method of selecting barley for malting will surely be developed in the future. The usual routine methods now available for evaluation of malting barley according to nitrogen content, germinative capacity, etc., restrict assessment to factors favouring high extract and good modification.

Malt. — In the malting process it is impossible to influence the β-globulins, and so nothing can be done at this stage to influence the stability of the resulting beer. Thus, what is regarded as a good malt is a malt which is as good as possible in all other respects, such as yield of extract, colour, flavour and head retention in the resulting beer.

With regard to beer stability any normally modified malt will do, and no special claims are justifiable for, e.g., soluble nitrogen figures (Kolbach’s figure). It is sometimes maintained, as recently by Mendlik,¹⁴ that it is important to have at least 40% soluble nitrogen in a malt with moderate nitrogen content. This demand might be valuable from other points of view, but no evidence is at hand to justify any special percentage of soluble nitrogen in malt in order to be able to produce a stable beer from this malt.

Mashing and Brewing. — Until recently, it has not been possible to influence the stability of beer to any notable extent during the mashing process. A large number of experiments have been carried out in various breweries, with the intention of improving the beer stability, all hitherto with very poor results. However, in 1949, Sandegren¹³ gave an account of some trial brews, carried out at the Stockholm Breweries, which for the first time gave results of practical value. These experiments were based on the fact that β-globulin has its isoelectric point at pH 4·9. Thus, the solubility of this globulin is at its lowest at pH 4·9, whereas the barley albumin which is valuable for head retention has its isoelectric point at pH 5·8.

In a two-mash decoction brew, lactic acid was added to the mash just after the second mashing and conversion. The pH was 4·9 and a short rest was kept at 85° C. (185° R). The mash was filtered in a mash filter, after which the normal procedure was followed.

A control brew was made without acidification. The stability of the resulting beers is shown in Table I.

It appears that this method of acidification can produce a beer which is considerably more stable, as far as chill haze and oxidation haze are concerned, than the control brew. At the same time the head retention was slightly improved. It will be noted that the pH of the experimental beer was only slightly lower than that of the control beer.

The present author has tried this method in some breweries and has been able to confirm Sandegren’s results. The influence of the acidification on the other properties of the beer, such as taste, favour and colour is more or less marked. The colour becomes comparatively pale when the pH of the wort is so low during boiling. This may easily be regulated by correcting the colour of the malt. Furthermore, pale lager beer produced from wort with a low pH (e.g., between 6·0 and 5·2) does not develop the reddish, or foxy, colour as do beers derived from worts with a higher pH. The taste of the beer from low pH worts is cleaner and the bitterness, especially, is fine and without harshness. On the other hand, rather more hops are needed in order to obtain the same degree of bitterness as in the beers produced from worts with a higher pH.

Wort cooling and aeration. — Beer stability flavour and bitterness are also influenced by the method of cooling and aerating the wort. In a number of breweries, where closed plate coolers have been installed, a certain improvement of beer stability has been observed. Gerard¹⁶ published some figures for indicator time-test (I.T.T.) measurements in wort before and after cooling in the old Baudelot cooling system, and after the introduction of the plate cooler and wort centrifuge.

From Gerard’s figures it appears that considerably lower I.T.T. values are obtained with the new cooling and clarification system than were obtained by the older methods. The same observations have been made in other breweries, thus confirming Gerard’s findings.

The lower I.T.T. value in the wort shows that the oxidizable substances in the wort have been oxidized to a lesser degree than was the case with the old method of cooling. Since the first substances to be oxidized in the wort are the reductones, we may assume that a larger part of these protecting com pounds is retained in the wort and the beer. This might explain the cause of the higher stability in the beer produced in breweries where the new system is employed.

Fermentation. — In the course of the fermentation there is nothing special that can be done to improve the stability of the resulting beer. The usual good brewing practice is of course always important and should be foll owed. Any abnormalities such as sluggish fermentations or infection of the wort or yeast are most likely to affect the bio logical stability of the finished beer. At the time of tanking or racking it is important to get the beer into the tanks with the least possible amount of yeast and air. Sufficient fermentable extract should be left in order to achieve a lively after-fermentation and a high CO₂ content.

Storage. — The temperature during the first weeks of the lagering period should be such as to promote a lively after-fermentation. Thereupon the temperature should be gradually lowered to below freezing point and kept there for at least two weeks in order to get the chill-haze substance to form and sediment.

Filtration. — In order to remove as much as possible of the chill-haze substance which has been formed in the cool beer, the nitration should be carried out at the lowest temperature possible in order to avoid re-dissolution of chill haze. Pulp filters having a considerable adsorptive effect are better than the modern sheet filters in this respect. The sheet filters should never be used alone for filtration of beer which is to be pasteurized and which must remain stable for any length of time. Sheet filters are appropriate when unpasteurized bottled beer is the main product and non-biological turbidities are out of the question. Kieselguhr filtration is a good method for pre-filtration, as the diatomaceous earth is a fairly good adsorbent of chill haze. These filters have their proper place in large breweries because of their big capacities.

Kieselguhr filtration is followed by a filtration through pulp filters or sheet filters for the final polishing. The combination of kieselguhr filter and sheet filter in particular has many advantages.

Centrifuging of beer is not advisable when a high degree of stability is demanded. This was pointed out by the present author in 1947,¹⁶ the reason being that the beer is heated slightly while running through the centrifugal bowl. This has a detrimental effect upon the beer stability, and centrifuged beer shows a higher degree of chill haze and oxidation haze than pulp-filtered beer. When special precautions are taken, such as chilling the centrifuge by direct expansion of liquid CO₂ so that the heating effect on the beer is avoided, centrifuging may give very satisfactory clarification. This was shown by Sandegren¹⁷ in 1949. With a specially constructed centrifuge, he was able to achieve very favourable results also. For practical use, however, such centrifuges are not yet on the market, but it is to be hoped that centrifuge designers will solve this problem since centrifuges have so many advantages, especially for small and medium-sized breweries.

Avoidance of air absorption. — The harmful effect of oxygen on beer stability will not be treated at length; this effect should be common knowledge to all brewers to-day. It is, however, surprising to see that precautions against aeration of beer are still often neglected. A stable pasteurized beer cannot be produced unless introduction of air from the time of fermentation until the bottle is closed is avoided, or at least kept at a minimum.

Chill-proofing methods. — As is well known, a series of methods is available for improving beer stability. For production of bottled pasteurized beer to be consumed on the home market (i.e., with a fairly quick turnover), it is not normally necessary to use artificial means. A sufficiently stable product can be made when all the precautions previously mentioned have been observed. When, on the other hand, the beer is to be sold on a more extended market it is necessary to adopt special treatments. This is, of course, especially important if the beer is to be transported over long distances through warm climates, as in the case of export beers.

Two different methods are used: (a) the chill haze substance is removed by precipitation by tannin or it is removed by adsorbents, e.g., Bentonite; (b) the protein part of the chill haze substance is decomposed by means of proteolytic enzymes.

It is rather difficult to carry out these treatments without loss of valuable qualities in the beer such as palate-fullness and head retention. A proper employment of either or both of these methods will always lead to a good result if careful attention has been paid during the brewing to the important points dealt with in the survey given above.

Pasteurization. — The pasteurization process in itself is practically without harmful effects on the beer if the latter is properly prepared with a view to its ultimate pasteurization. Above all other precautions to be taken in this respect is the avoidance of air (oxygen). A bottle containing about one third of a litre of beer for pasteurization should never contain more than 3 ml. of total air as a maximum.

The Foam
The two main properties of interest in beer foam are head formation and head retention. In a particular beer possessing a substantially normal head-retaining property, the head formation is mainly a function of the CO₂. content of the beer (Helm and Richardt¹⁸), and it is quite clear that no head formation could be accomplished, if no head-retaining properties were present. In bottled beer there is a tendency to increase the gas content, which is no doubt one of the many American influences experienced since the war. This is done in the hope of camouflaging deficient head retention with an overwhelming head formation. It might help a little, but not very much. There can, however, be no doubt that high gas contents are helpful with regard to non-biological stability, as high CO₂ contents in beer diminish the risk of air absorption.

Head retention has been studied a great deal, but not with the same success as have the non-biological turbidities. More is known about how to destroy the head than about how to improve it.

The fundamental principles for head retention are no doubt present in the barley, and individual parcels of barley may contain more or less of these constituents. In the malting process these principles can certainly be influenced. Head retention in the resulting beer is decreased with increasing germination periods and increased with increasing curing temperatures (Helm¹⁹). Unfortunately, it is not known with certainty which are the important compounds and what happens to them in the process of malting, but proteins, other surface-active compounds, and carbohydrates are undoubtedly playing their parts.

With regard to proteins, the classical theory was that certain intermediate protein compounds, the so-called proteoses, were valuable foam-stabilizing particles, and in order to produce these in quantity the so-called proteolysis rest was recommended in the decoction process. In later years, however, quite the opposite point of view has been maintained. The writer has experienced very satisfactory results with mashing methods in which proteolysis was avoided.

The protein particles which concentrate in the foam bubbles and may be found in collapsed foam have a molecular weight of about 10,000, as Sandegren^6,7 has shown. He has also shown that these proteins originate from barley albumin and are surface active. Since barley albumins are water-soluble, they are readily dissolved in the mashing process, and they do not need any breakdown. Accordingly, the proteolysis rest should be superfluous in mashing.

In a different way a strong proteolysis might also be detrimental. Increasing the amount of certain amino acids by proteolysis, will give a wort which contains too much fusel oil after the fermentation. Sandegren⁷ has shown that beer containing much fusel oil produces foam with lower head retention than beer containing little fusel oil. A certain surface tension is necessary, but this must be kept within a certain limit, since too low, as well as too high, a surface tension will have a bad effect; and it is also well known that addition of surface-active agents like higher alcohols, soap, and quaternary ammonium compounds will destroy the foam. As a certain variable amount of higher alcohols, i.e. fusel oils, is formed in the fermentation process, there is here one of the factors which should be studied with a view to practice.

In addition to proteins and other known surface-active agents, it should be mentioned that some hop constituents, too, are surface active and do concentrate in the foam. The amount of hops used in brewing, however, does not seem to be so important with regard to head retention.

Little is known about the part played by carbohydrates in head retention. If any relation exists the materials concerned should be dextrins and pentosans. Such materials may increase the viscosity, which is of some influence on the head retention and fullness of the beer; viscosity is determined not so much by proteins as by carbohydrates. Short-grown or moderately modified malt produces wort with higher viscosity than normal and over-modified malts (Helm¹⁹); at the same time, the beer resulting from the former malts has the better head retention. It is not known whether the carbohydrates which influence viscosity are dextrins or pentosans. Some indications are available (Piratzky and Wiecha²⁰) that a somewhat obscure carbohydrate, called amylan by O’Sullivan,²¹ may play an important part in this respect.

There is a great want of new methods for examination of malt and wort. The usual malt analysis for instance, tells about extract and colour, but very little about the real quality of the malt, and in particular nothing at all about the quality of the beer that can be made from it.

An attempt was made to study whether any relationship could be found between a protein fraction in the wort on the one hand and head retention in the beer on the other.

It seemed likely that a suitable protein fraction would show such a relationship, since proteins are certainly related to head retention. Formol titration and tannin precipitation have formerly been used, but without success. So, a new method was chosen, viz., Lundin’s precipitation²² with phosphomolybdic acid. By this method 30-40% of total nitrogen is precipitated (PM fraction), and a certain limit is obtained which might be assumed to fall within the region of the above-mentioned proteoses of intermediate molecular size. A high PM fraction would mean a low degree of break-down of the proteins and possibly a high proportion of proteoses and less free amino acids. Consequently, a good head retention might be expected.

Thirty-six samples of brewhouse worts and the corresponding beers were analysed from a number of lager breweries. In a majority of cases the PM fraction fell within 32 and 38% of total nitrogen, and no relationship could be found between these figures and head retention measured by Blom’s method. The head retention as half-life period in seconds ranged from 82 to 99 sec. (90 sec. corresponds to normal head retention in lager beer). If, however, the PM fraction went below 30% (30-26), the head retention in the resulting beer was found to be abnormally low (61-81 sec). It is, however, most likely that other factors have played a part in these abnormal cases, since they all came from the same brewery. Hence, this attempt gave a negative result, and so it will be necessary to try a new path.

The problem of head retention is a most complicated one, and a number of unknown factors play their parts. Of purely practical things the after-fermentation might be mentioned, as a lively after-fermentation during the storage period in the lager process and the conditioning in the ale system is most important for the foaming properties of the beer. In this process scientific research should be made.

Relation between Head Retention and Stability
It has often been maintained that it is not possible to produce a stable beer which shows at the same time a good head retention. It was assumed that the same proteins which were important for making the head stable, were the cause of non-biological turbidities. It is now known that this is not the case. Sandegren⁶ has definitely shown that the substance which produces chill haze has no effect whatsoever on head retention. He has shown that barley albumin produces the nitrogen compounds which are important for foam stability. He has studied the proteins which are concentrated in the foam bubbles, and shown that they are of a molecular weight of about 10,000, whereas the haze forming proteins are globulin particles with a molecular weight of about 30,000-40,000.

When proteolytic enzymes are used in order to improve non-biological stability in beer, both globulins and albumins are broken down at the same time, and consequently both turbidity and head retention are affected. But this parallel effect cannot, of course, be any proof of identity. The same explanation holds good when other methods are used for stabilization of beer, e.g., sharp nitration or tannin-treatment. Consequently, future possibilities are to be found in selective treatment, i.e., methods which will only affect the haze-forming globulin-tannin compound, leaving the albumins untouched.

As already mentioned, there is very little which can be done during malting in order to influence the globulin-tannin contained in the barley and malt with a view to influencing non-biological stability. At this point, viz., the malting, there are on the other hand possibilities for influencing head retention in the resulting beer, as has already been pointed out.

In conclusion it may be said that theoretically there is nothing to prevent the production of a beer which is at the same time almost completely stable non-biologically and yet produces a very stable head when poured out. But for the time being it remains an art which only a lucky and clever brewer can master. It is a future aim to make it an applied science to produce this ideal product.

Bibliography
1. de Clerck, J., this Journ., 1934, 407.
2. Mendlik, F., Rev. int. Brass, et Mali., 1940, 154.
3. Wiles, A. E., this Journ., 1950, 183; Brewers’ Guild J., 1950, 36, 141.
4. Helm, E., this Journ., 1936, 379.
5. Sandegren, E., Proc. Eur. Brew. Conv., Scheveningen, 1947, 33.
6. Danielsson, C. E., and Sandegren, E., Acta chem. scand., 1947, 1, 917.
7. Sandegren, E., Proc. Eur. Brew. Conv., Lucerne, 1949, 78.
8. Quensel, O., Dissertation, Stockholm, 1942.
9. Hartong, B. D., Woch. Brau., 1937, 54, 34.
10. Helm, E., this Journ., 1939, 80.
11. Hartong, B. D.. Woch. Brau., 1934, 51, 409.
12. Saverborn, S., Danielsson, C. E., and Svedberg, T., Svensk hem. Tidskr., 1944. 56, 75.
13. Sandegren, E., 1st int. Cong. Biochem., Cambridge, 1949, 645.
14. Mendlik, F., this Journ., 1950, 134.
15. Gerard, F., Bull. Ass. Ec. Brass. Louvain, 1949, 45, 149; this Journ., 1950, 219.
16. Helm, E., Proc. Eur. Brew. Conv., Scheveningen, 1947, 141.
17. Sandegren, E., Svensk Bryggeritidshr., 1949, 112; this Journ., 1950, 337.
18. Helm, E., and Richardt, O. C, this Journ., 1936, 191.
19. Helm, E., Wallerstein Lab. Commun., 1939, 2, 40.
20. Piratzky, W., and Wiecha, G., Woch. Brau., 1938, 55, 97.
21. O’Sullivan, C, J. Chem. Soc, 1882, 41, 24.
22. Lundin, H., Die brautech. Untersuchungsmethoden, Pawlowski-Doemens. Niirnberg: Verlag Hans Carl, 1947; p. 161