[12/16/15 Here is another paper by Percival Greg: Fermentation in Rum Distilleries.]
This post contains four blockbuster papers from Percival H. Greg who is, so far as I can tell, unknown in modern historical reflections on rum. Greg gave Jamaican rum its largest quantum leap in identity with a state of the art study of yeast strains and promotion of top fermenting yeast no. 18 to maximize heavy rum aroma. This all happened as Caribbean rums were hit with hard times. As evidenced in another post, all the Caribbean pretty much watched and then eventually adopted the process. Before the breakthroughs of Greg, islands like Trinidad were only able to produce “common clean” rum but could not compete with European advances in sugar beet rums. Unfortunately, I’m not giving you any kind of a synthesis. This is a primary document, but luckily Greg’s writing style is simple and educational. You are on your own and I suspect people will only end up here by google search terms.
It takes a lot of work to extract these from a PDF and I’m quite busy lately so I’ll let someone else take a turn at annotating them.
JAMAICA BULLETIN OF THE BOTANICAL DEPARTMENT.
New Series. MARCH, 1895.
By Percival H. Greg.
I do not think I am wrong in saying, that the smell of rum, really good rum that is, is one of the most delicious scents that can be imagined. There is in addition something so peculiar and undefinable about it; it is so different from the smell of any other spirit that the more we smell it, the more we are puzzled to say to what its aroma is really due.
Rum like almost every product of commercial importance has been analysed. The most important work in this direction has been done by foreigners, and those chiefly Germans, who have lately been making a great effort, and will probably continue doing so, to produce a spirit from Beetroot juice, molasses, or cane sugar, to vie with Jamaica rum. So far I am glad to note, these efforts have been unsuccessful, but Jamaica should remember the painful lesson taught by the rise and growth of the Beet-sugar industry, and should take a leaf out of their adversary’s book, in not disdaining to study science in connection with their manufactures.
Up to now analysis of rum, and this remark applies with equal force to brandy and arrack, has been rather barren of results. One of the first aims of analysis of any commercial product, has always been to determine what are the normal constituents of the article in a state of purity, and in what proportions these constituents are present. Unfortunately the difficulties in the way of this are manifold. No chemical analysis so far can tell us, whether a rum has been adulterated or not. In this respect an opinion emanating from an experienced rum dealer, is worth far more than that of the analyst. This however is not so surprising as might at first sight appear. Rum is not a definite chemical product, we cannot write the formula of rum, and any one of its constituents may vary in the proportion in which it is present, or indeed some may be absent altogether, without our being able to say, that such a change is due to artificial manipulation. The great obstacles in the way of analysis of rum, brandy, and arrack, are the large quantities required for analysis, and the consequent costliness of the operation; the uncertainty as to whether the spirit, when this is forthcoming in sufficient quantity, has not been adulterated either at the place of its production, or in its subsequent passage through, other hands; the fact that all spirits undergo chemical changes during storage; and the fact that the particular substances constituting the aroma are present is such infinitesimally small quantities.
The two most important works on the analysis of rum, brandy and arrack are those of Dr. E. K. Windisch and Dr. Eugene Sell, members of the Kaiserlichen Gesundheit’s Amt. Both of these works embrace the results of previous knowledge on this subject, and would I am sure, be full of interest to those distillers who take a scientific interest in the manufacture of rum.
The chief points touched upon in the analysis of rum are the percentage of alcohol, reckoned as ethyl alcohol, colour, taste, smell, reaction, whether acid or neutral ; presence of aldehyde and higher alcohols, percentage of free acids and ethers. The presence of methyl alcohol as a normal constituent of genuine rum, is up to the present time a disputed point.
The following tables are taken from Dr. Eugene Sell’s book, Ueber Cognak, Rum and Arak (concerning Cognac, Rum and Arrack) :—
On looking at the table, the percentage of Alcohol will be seen to be fairly uniform, the greatest difference discernible being between No. 5 Jamaica at 79.06 vol. per cent, and No. 11 Cuba at 73.73 vol. per cent.
As regards colour, smell, and reaction whether acid or neutral, nine samples from Jamaica had a reddish brown colour, the two from Demerara might be called black-brown, and the three from Cuba bright yellow. The smell of the various samples was of course different but they all possessed in a high degree the aromatic smell of rum. All the samples reacted acid.
Aldehyde. — In every case the presence of aldehyde was proved by means of the metaphenylenedianine, as well as with fuchsine-sulphurous acid reaction; with aniline and hydrochloric acid the presence of furfurol (furfurane aldehyd) was established.
Free Acids.The following table shows the figures obtained:—
The Ethers. [Esters]
These are considered to be the chief source of the aroma of rum, and as some of my readers may not know what are the nature of these bodies, and may perhaps think I am speaking of the ether of the B.P. which is used in medicine, it may be well to offer some explanation. The ethers of which I am about to speak are generally known as the “fruit ethers” so called because they possess a pleasant fruity smell. These ethers are formed by the chemical combination of an alcohol, generally ordinary ethyl alcohol with an acid, generally an acid belonging to the order of the carbon compounds. Everybody is familiar with the smell of pine-apple rum. This is due to the presence in the rum of a minute quantity of butyric ether or ethyl-butyrate, which is a combination of butyric acid with ethyl alcohol. Another ether almost invariably found in rum is acetic ether, a combination of acetic acid with ethyl alcohol, which has a smell of fermented apples. The other ethers most commonly met with in rum are formic and capric ethers. The following table shows the figures obtained in the estimation of the ethers in the fourteen samples specified :— [Table is in the link]
Most writers have attributed the aroma of rum to the presence of Butyric ether, indeed Gaber in his book “Die Liqueur Fabrikation” (Liqueur Manufacture) states that it is only necessary to allow cane sugar molasses to ferment in Europe at a temperature of 40° C. a temperature favourable to the production of butyric acid, and therefore a necessary preliminary condition to the formation of butyric ether, in order to produce a spirit, which after sufficient storage cannot be distinguished from a true rum. But if we glance at the tables we notice that acetic ether is present in far greater quantity than butyric ether, in one case the proportion or acetic ether to butyric is over 200 : 1, and we must therefore conclude that it is to the comparatively large amount of the first named ether, that rum owes its characteristic aroma, in as far as the characteristic aroma is derivable from the presence of ethers. Now in considering these figures from a practical and commercial point of view, we should like to know, what is the influence of the presence of these ethers on the commercial value of the rum : that is, is an increment of ethers present in the sample followed by a corresponding increment in the market price, and vice-versa? On this point the figures given in the preceding table can give us no information, as the money values of the rums are not stated; but in Windisch’s “Brandy Analysis,” I came across some analysis of rum which possess a greater significance for us, inasmuch as the prices of the rums are also given. In following out this idea I selected from a considerable number of examples four expensive and four cheap rums, and then set myself to compare them together with special regard to the quantities of acids and ethers present. In the original work the prices are expressed in German coinage and for the measure of one litre, but in order to make the figures more intelligible to my readers, I have translated them into the corresponding values in English money per gallon.
On examining these tables we learn that the expensive rums as a class are richer than the cheap rums in the total amounts of acetic and capric acids, and poorer in formic and butyric acids while the table of ethers shews us that the expensive rums as a class are richer in the total amounts of all the four fruit ethers. It would be very rash however to jump to the conclusion, that the richer a rum was in fruit ethers the higher would be the price which it would command ; a glance at the total amount of fruit ethers present in each of the expensive rums will show that the rum at nine shillings and eleven pence, is poorer in the total quality of fruit ethers than two of the rums at nine and two pence and eight and seven pence respectively. But on the other hand we may observe in the case of cheap rums, that the rum at four shillings is richer in the total quantity of ethers, than the rums at three and two pence, and two and eight pence. In order to explain this apparent contradiction we must remember that the fruit ethers though contributing a great deal, do not contribute everything to the aroma, and also that even if they did, that it is quite possible to have too much of a good thing. It has been found for instance, that by the mere addition of fruity ethers to a neutral spirit, it is not possible to imitate exactly the aroma of a genuine rum. If this were so, in all probability, Jamaica rum manufactured in Jamaica, would belong to the glories of the past! A spirit so treated will have a too pronounced and penetrating odour and a sharp burning taste. “The flavour of a true rum on the contrary is always soft and mild, the aroma appears to be to some extent ‘covered,’ and leaves a kind of oily impression upon the tongue” (Windisch). As regards the other aromatic constituents of rum there remains, the higher alcohols generally grouped together under the name of fusel oil, the organic bases, the essential oil of rum, and a fatty acid of a fruity nature. Fusel oil has generally been identified with amyl alcohol (Iso-butyl carbinol), which possesses poisonous properties, and is of a disagreeable smell, but since higher alcohols do not necessarily mean amyl alcohols, it would be unwise in the present state of our knowledge concerning the higher alcohols of rum to draw any inference as to their influence on the aroma. For the present we merely give the results arrived at in the determination of the higher alcohols, without comment.
Fusel Oil. [Missing table linked here]
The Organic Bases
Lindet (Comp. rend, de I’Acad. des Sciences, 106, p. 280) has found that different kinds of rum from Rénunion, Guadeloupe and Martinique, are especially rich in organic bases, which he considers to have been formed in the molasses by certain micro-organisms, previous to fermentation (Sell). The organic bases are evil smelling substances of a poisonous nature, but according to my experiments which I shall communicate at another time, I find they exist in large quantities in the skimmings, and I therefore think that the supposition of Lindet, as to the intervention of micro-organisms in their formation, is not correct.
The Essential Oil of Rum.
In the estimation of the higher alcohols, which is based on the absorbtive capacity of chloroform for these substances, Dr. Windisch observed both in the case of arrack and rum, that after the chloroform which had been shaken up with the rum was evaporated off, a small drop of an unsaponifiable terpene kind of oil remained behind, which possessed in a remarkable degree the aroma of rum or arrack as the case might be. Two years ago, while performing some analyses of rum in Germany, I repeated the chloroform experiment and having found the same substance, I extracted several portions of rum in the requisite manner, added the residue to another portion of the same, and sent this sample along with another sample of the same rum, but untreated, to a rum seller in London. The report of the expert on the two samples was, that the sample to which the addition had been made, of which addition it is needless to say he was ignorant, was worth about threepence a gallon more than the untreated sample of the some rum.
The Fruity Acid.
Herzfeld in his analysis of rum (Dr. Alex. Herzfeld, Versuche zûr Darstellûng Rum-artiger Produkte). (The Manufacture of Rum Products) discovered the existence of an acid of a fruity smell, but the materials at hand were not sufficient to enable him to identify it. It is evident that the presence of this acid in greater or less quantity would have a corresponding effect on the quality and therefore on the price of the rum, and I think that I shall presently be able to give some interesting information as to its occurrence.
With this we conclude the enumeration of the bouquet-producing constituents in rum, which up to now have been discovered by analysis. It seems evident that the points to be aimed at by those desirous of improving the quality of their product, are the production of fruit ethers, the oil of rum, and the fruity acid in the resulting spirit.
A CONTRIBUTION TO THE STUDY OF THE PRODUCTION OF THE AROMA IN RUM.
By Percival H. Greg.
This account of some experiments which I have been making in this direction, and which I venture to submit to readers of the Bulletin in general, and to Planters and Distillers in particular, pretends to be simply what it is entitled—a contribution—the question indeed is a wide one and may be approached from many sides.
The aroma of rum may be said to be mainly due to five causes, 1st the nature of the plant from which rum is made, the sugar cane; 2nd the soil on which it is grown ; 3rd the fermentation; 4th the distillation ; and 5th the storage in cask, at the high temperature prevalent in the tropics. My researches as will be seen, deal exclusively with the aroma developed during fermentation, with special regard to the influence exercised in this direction by a particular variety of a peculiar type of yeast. They were suggested to me by the brilliant results attained by Prof. Hansen in his employment of pure yeast in the manufacture of beer. Following out Prof. Hansen’s ideas as to the great influence exercised by the type of yeast on the nature and resulting products of the fermentation, my work at first consisted in isolating and cultivating pure, according to Prof. Hansen’s method, as many different varieties and species of yeast, as I could obtain from the materials, molasses and dunder, sent to me from Jamaica, and instituting with such cultures trial fermentations on a small scale. I gave a preliminary account of these researches in the “Sugar Cane” of Nov. 1893, in which I stated that I had isolated a considerable number of varieties of Jamaica yeast, possessed of very different properties.
One yeast in particular seems to me to have a special bearing on the production of the aroma in rum. It belongs to the type known as “top fermentation” yeast, i. e., it throws up a “head” on the surface of the fermenting liquid, which, in molasses and dunder, is of a beautiful golden colour and very tenacious in character. The progress of the fermentation is a slow one, varying, according to the composition, concentration and temperature at which the wash is fermented, from 10 to 14 days. The fermentation of the liquor is a very quiet one, the gas being given off slowly in small bubbles, and at some stages fermentation is hardly noticeable. During fermentation, although there is a somewhat “fruity” smell, a definite aroma cannot be said to be produced, but after the fermentation is concluded, if the liquor be allowed to remain quiet, say from 24 to 36 hrs. a delicious aroma can be distinguished. In order to prove without a doubt, that the aroma produced was due to this germ, the following experiment was performed. A certain quantity of molasses and dunder and water, mixed together in suitable proportions, was taken and sterilised by boiling. It was then allowed to cool in contact with air previously freed from all germs, and when a sufficient amount of air had been absorbed the liquid was equally divided between two fermenting cylinders which had also been previously sterilised. One cylinder was set in fermentation by means of this particular yeast, which I call No. 18, and the other cylinder was fermented by another Jamaica yeast which I will call No. 4. The two cylinders were then placed under exactly the same external conditions, and fermentation allowed to proceed. The appearance of the two cylinders during fermentation was characteristic. No. 18 was covered with a thick golden buttery head and fermentation was slow, while in No. 4 cylinder the yeast remained entirely at the bottom, and the fermentation was rapid, and was what is technically called a “champagne” fermentation. At the end of 5 days fermentation was entirely at an end in No. 4 cylinder, while in No. 18 it was still in progress. No 4 cylinder was allowed to stand 36 hrs. No aroma was developed. Eventually fermentation was finished in No. 18 in 12 days, and the wash allowed to stand 36 hrs.,—a heavy fruity aroma was developed. This experiment clearly shows that the aroma in question was due to the influence of No. 18 yeast, since the two washes fermented were identical in composition, and were fermented under exactly the same external conditions. It raises too a point of some practical importance to which I would call the attention of estate owners and distillers. There seems to be a general unanimity of opinion among planters that in order to produce a fine rum, the “wash” must be allowed to “die down” thoroughly. To accomplish this however necessitates in many cases building larger still houses, which many estates in these hard times are unable to do. But does it not seem evident from my experiment that the amount of benefit to be derived from the enlarging of the still house and thus giving the liquor room and time to attenuate thoroughly will vary very much according to whether an aroma or non-aroma producing yeast has the mastery in the vats? So far we see the problem must be approached from two sides. The case however presents other points of interest. How far in the experiment under discussion was the aroma due to No. 18 yeast? Did the yeast excrete the aroma, so to speak, or did it form it from, or by transforming, certain substances in the liquid? In order to settle this question, I fermented separately, by means of No. 18 yeast, refined cane sugar, dextrose, cane juice, and molasses, leaving the liquor to stand 36 hours after the completion of fermentation : in no case was the aroma developed. I repeated these experiments, but the results were the same. I need hardly add, that where necessary, yeast nutriment was added in order to produce a normal fermentation. This puzzled me for some time. Dunder and molasses and No. 18 yeast gave the aroma, but molasses and nutrient salts and No. 18 yeast did not : therefore the aroma must have been produced from No. 18 yeast acting on some substances in the dunder. But dunder is simply the residue of wash which has been previously fermented and distilled, and is in fact the residue of cane juice, and skimmings and molasses. But neither cane juice, which contains those substances which eventually go to form skimmings, nor molasses, gave the aroma. Perhaps then it was due to the process of boiling in the still? Accordingly pure cane juice and molasses were allowed to undergo fermentation and were then distilled, and fresh wash set up with the resulting dunder, but no aroma was developed: it was evident then that dunder, as dunder, had nothing to do with the formation of the aroma in question. I must here make a short digression. While I was engaged in Europe in isolating different yeasts from the materials (molasses and dunder) sent to me from Jamaica, I searched for a long time in vain for a yeast capable of producing a definite aroma. With this object I must have made certainly not less than two hundred pure cultivations. One yeast only attracted my attention as seeming to be able to produce a faint aroma, certainly more than the rest. This yeast therefore I examined more closely. On testing the mixture of dunder and molasses which I was fermenting with this yeast, it was found to be exceedingly acid. It was thought therefore that such a great acidity was injurious and might probably interfere with and prevent the yeast from exercising its physiological functions to the full. I accordingly partly neutralised the acidity of the dunder with a few drops of caustic soda, and put the liquid which No. 18 yeast had nearly finished fermenting, away to stand. After standing for about three days and when fermentation was at an end the characteristic aroma was developed. This yeast I afterwards named No. 18 and is the one used in these experiments. I had not time then to proceed any further with the question, and remained satisfied with the explanation I have adduced. When however my experiments in Jamaica led me to see that the aroma could not be produced from fresh cane juice or molasses, or even from cane juice and untreated skimmings, or from molasses and untreated skimmings, but yet could be produced by the help of the dunder acted upon by No. 18 yeast and bearing in mind my former experiment in Europe, of partially neutralizing the dunder, which had resulted in the production of the aroma, I bethought me of the treatment which the cane juice undergoes in the boiling house.
This as a general rule in Jamaica consists in treating the cane juice with caustic lime. I determined therefore to try the experiment of fermenting cane juice which had undergone this treatment. But as an alkaline medium is unfavourable to alcoholic fermentation, and as the alkaline skimmings from the boiling house are brought down, whenever possible, on acid dunder, I determined in order to test the efficacy of the treatment and at the same time to provide a favourable fermenting medium for the yeast, to neutralise the alkalinity of the treated cane juice with sulphuric acid, using a sufficient excess to produce a slight acidity : dunder of course could not be used in this experiment. A portion of fresh cane juice was therefore taken and divided into three parts I, II and III. No. I was made alkaline with caustic lime, No. II was not treated and served merely as control experiment, while No. Ill was not treated with temper lime, but was faintly acidulated with sulphuric acid. This last also served as a control experiment, as it might be argued, that should any aroma be produced in No. I sample, it might have been caused, not by the treatment with an alkali, but to the subsequent liberation of aromatic vegetable acids, or volatile vegetable acids, capable of forming fruit ethers perhaps by the stronger acid sulphuric. I must add that in order to preclude the possibility of action of germs other than No. 18 yeast, which would of course be naturally present in the cane juice, that the three portions of liquid were sterilised before being fermented. No. I. was sterilised (boiled) after the treatment with lime, in order to imitate the treatment to which the skimmings are subjected in the boiling house, and the sulphuric acid added just before the fermentation. No. III. was boiled before the treatment with sulphuric acid in order that if volatile acid should be liberated that they should not be driven off by the heating, that is that the maximum effect if any, due to their presence, should be obtained. No. II. sample was of course simply boiled without any addition. After these three samples had been boiled and allowed to stand in contact with sterilised air for a sufficient length of time, an equal quantity of No. 18 yeast, in a state of absolute purity was added to each portion and the liquid allowed to ferment, the outside temperature being the same in each case. In No. I. sample, which had undergone the treatment with lime, the characteristic aroma was developed. In samples II. and III. the aroma was not developed. It is but fair to state that the aroma produced was not very strong, but there was no denying its presence. Here then we have the four factors necessary for the production of the aroma in question. First the germ, No. 18 yeast; secondly, the medium skimmings or cane juice; thirdly, the treatment of the liquid —heating with caustic lime, or caustic alkali; fourthly,the question of time—the wash must stand 24 to 36 hours after fermentation has been completed. Be it borne in mind that the absence of any one of these 4 factors, will result in the non-production of the aroma in question. I expressly state here of the “aroma in question” because it is not contended that no other aroma can be produced by any other germ or germs which may be active in the fermentations. Indeed the treatment with lime for reasons which I shall show at another time, has an effect upon the flavour of the resulting spirit, independent of the action of any particular germ. In order however to make quite sure that the activity of 18 yeast is necessary in treated cane juice to ensure the production of the aroma, treated cane juice was fermented by several other Jamaica yeasts, but no aroma was produced. My researches hitherto have been directed to show that a certain aroma can be produced by a particular type or variety of yeast, which cannot be produced by other germs in my possession, but that though the activity of this yeast is essential, certain other conditions are equally essential. In other words the aroma produced during fermentation is the resultant of more than one force, but that the type of yeast employed plays a very important part in the matter and the probability is that this remark applies to any other aroma produced during fermentation. I have in this case been able to isolate and identify these forces, which is advantageous as showing how No. 18 yeast may be employed to produce the aroma, with its maximum effect. But it will be seen that so far I have touched on the matter somewhat superficially. I have yet to show what is the substance or substances which are acted upon by the caustic lime in the cane juice and the way in which No. 18 yeast acts upon them to produce an aroma, and I have yet to show why the effect of the lime is not rendered nugatory by the after addition of sulphuric acid. This however would make my paper too long and as I am still engaged in investigations on these points, I must defer any explanation for the present. One thing however further experiments have taught me with absolute certainty, that in order that the treatment with lime be efficacious in its influence on the flavour of the rum—and this applies equally, whether 18 yeast is used or not—the skimmings must be thoroughly heated in the syphons after the treatment, with lime.
THE JAMAICA YEASTS.
By Percival H. Greg.
In a collection of papers from the Demerara Argosy entitled if I remember rightly the “Planters Manual” 1889 [I think he miss-remembers and its actually the Overseers Manual which isn’t digitized to my knowledge], there is a very interesting article on “How to make German Rum, by a Jamaica Distiller.” Among other things the author mentions that the liquor throws up a thick golden head, that fermentation is very slow, and that no particular characteristic aroma is produced until after fermentation has been concluded. This corresponds so exactly with the behaviour of my No. 18 yeast during fermentation that I am inclined to think that the yeast forming the golden head or “Rum fat” as he describes it, is the one which I call No. 18. The author after stating minutely the methods to be employed in the manufacture of this German Rum, confesses that this recipe is not always attended with successful results, inasmuch as that some estates, trying all they can, never produce German Rum, while other estates produce it without any apparent effort. Very interesting it would have been if the author had stated, which as far as my recollection serves me he did not, how much importance he attached to the presence of this “rum fat” in producing the aroma, and as to whether this characteristic fermentation was absent or at least not permanent in those estates which tried to produce German rum, and failed. There would be nothing very startling if this were so. All the most recent researches go to show that the influence exercised by the particular organism active in the fermentations on the flavour of the resulting aroma of the Beer, Wine or Spirit has up to within recent years been in many cases under-estimated or indeed not taken into account at all. I have seen an organism which out of pure sugar was able to produce liquor which smelt like pure pineapple essence, and I have in my possession, two varieties of the type Saccharomyces anomalus which produce a distinct pineapple flavour in molasses. Thus in Hansen’s Untersuchungen us der Praxis der Gährungsindustrie, which translated freely signifies “experiments in practical fermentation,” mention is made of the results attained by a Dr. Nathan in Rottweil in the use of selected types or varieties of yeasts in the preparation of fruit wines. The experiments were carried out on a large scale, and are therefore the more important. The conclusion to be drawn from them was that the quality and whole character of the fruit wines, is much more dependent on the character of the yeast which plays the leading part in the fermentation than is the case with grape juice. If (writes Nathan) I examine the 40 fermenting vats which I had filled with one and the same Must (fruit-juice) whether it was from berries or apples or pears, and then afterwards infected each with a different kind or type of yeast, the products of the fermentation differed from each other in such an extraordinary manner that no one would have believed that he had to do with one and the same material. While some types of wine yeast gave for example the apple-must a very pronounced winey taste and smell, others showed themselves able to alter the material but little. Some yeasts gave a very disagreeable after-taste to the must. Other examples could be given showing that the flavour of cream, butter, the ripening of cheese, the aroma of tobacco, etc., are due to the activity of special types or varieties of micro-organisms. Returning again to the subject of the Jamaica yeasts, there is another point to be discussed. In my last article I mentioned another fermentation which I obtained with a Jamaica yeast which I called No. 4. I showed that there were two apparent differences between the two yeasts, one the difference in the resulting products of fermentation, i.e., the aroma, and the other the time required by the two yeasts to ferment the same quantity of the same mixture of molasses and dunder, i.e., No. 4 requiring 4-5 days, and No. 18. 10-14 days. Here we see then that the kind of yeast employed is one of the deciding influences in what is a most important point in the Still House, viz., the question of time. It may not be out of place here to give a list of some of the Jamaica yeasts which I have isolated and proved in fermentations in my Laboratory.
Yeast No., Time of Fermentation, Attenuation of Wash., Alcohol. vol. per cent.
To translate Brix into Jamaica Saccharometer multiply by 1.33. Thus 21-5 Brix=27.9-6.55. In considering these figure they must be regarded in the light of a comparative rather than an absolute test of the capabilities of the various yeasts. In this experiment the yeasts were compared together under exactly the same conditions and therefore the differences shown can only be due to specific differences existing among the yeasts themselves. It is quite possible that the differences might become still more marked under different conditions such as for instance an increase in the initial density of the liquor. Thus with No. 17 I have under favourable conditions obtained an attenuation of 36.4-8=28.4 degrees attenuation (Jamaica Saccharometer) in from 5-6 days, but it does not necessarily follow that all the other yeast in my list would under those conditions give corresponding results. As it is, however, the differences shown in time of fermentation, amount of attenuation, and the quantity of alcohol produced, are worthy of attention. As regards the attenuation it must be noted that the Brix saccharometer was used instead of the one in use in Jamaica, and that if judged by the latter standard the number of degrees representing the attenuation would be greater. The amount of alcohol obtained is expressed in percentage, i.e., in the number of volumes of absolute alcohol present in 100 volumes of the wash. The greatest difference in the amount of alcohol produced is between either of yeasts, Nos. 1, 5, and 7, and No. 4. Thus on 100 gallons of wash the difference is 1 gallon of absolute alcohol—10 gallons per 1,000 gallons of wash. Other differences which are not indicated in the table were also observed among the yeasts. Thus some started fermentation quicker than others, some reproduced themselves more than others; and some formed a deposit which adhered tightly in a hard pasty mass to the bottom of the fermenting vessel, while others formed a deposit which was easily disturbed. With the exception of No. 18, and No. 19, which I have not included in the list, the yeasts were all of the “low” fermentation kind, i.e., remained at the bottom of the liquid. Nos. 18 and 19 are of the “top” fermentation kind i.e., throw up a “head” on the surface of the liquor during fermentation. With the exception of No. 18 the difference in the flavour of the resulting distillates was not very marked but the quantities operated upon were too small to enable a correct judgment to be formed, and of course the influence of storage in cask had to be left out of account. These results be it observed are results obtained in the Laboratory. If it be asked what results would be obtained by working with such and such a yeast in the Still House? The answer is that that this can be best determined by direct experiment in the Still House. There is however very little doubt that the comparative differences shown here would also obtain in the Still House. No. 18 yeast for instance will always be by comparison a much slower fermenting yeast than No. 1, and will produce a more aromatic spirit. While No. 4 will be sure to produce more alcohol from a given weight of sugar than Nos, 1, 5 and 7. The object of this paper has been to show that characteristic differences exist among Jamaica Yeasts which are active in the rum fermentation, that these differences are worthy of study, and may if placed under control be used with great advantage in the Still House. The principle which is advocated here is the selection by systematic experiment of that particular type or variety of yeast which is best suited for the kind of work it has to do and the cultivation and propagation of it in sufficient quantity for use on a commercial scale i.e., for fermentation in the Still House. If this were not possible our interest in the question would be confined to its scientific aspect, but pure selected types of yeast are now in use in large numbers of Breweries and Distilleries. There seems no reason then to doubt that the introduction of selected types of yeast into Distilleries here would also be attended with advantage and in my opinion this forms the basis of a solid improvement. At any rate this reform seems worthy of a thorough trial on a practical scale, and I append here a description of the apparatus by which the yeast desired may be grown absolutely pure in sufficient quantity for this purpose. I would however strongly advise all those who may take an interest in this subject to purchase “Micro-organisms and Fermentation” by Alfred Jörgensen published by F. W. Lyon, Eastcheap Buildings, London, a short review of which by me appeared in the “Bulletin” for May. Illustrations and descriptions of the two “Propagating” apparatus are given in it. The prices of the apparatus are approximately as follows:—
Apparatus Model Hansen and Kûhle, 1 sterilizing and one fermenting cylinder … 1,600 Krones.
Air pump and air chamber for propagating apparatus with stop valve, safety valve and Manometer … 800 Krones.
Apparatus Model Jorgensen and Berg. … 1,350 Krones.
The prices are given in Danish Currency, the value in English pounds sterling will be found approximately by dividing the number of Krones by 18. (18 Krones —20 shillings).
CONTRIBUTION TO THE STUDY OF THE PRODUCTION OF THE AROMA IN RUM.
By PERCIVAL H. GREG.
THE OIL OF RUM AND THE ORGANIC BASES.
Every Distiller knows that rum made from pure fresh cane juice is devoid of any pronounced flavour, as compared with rum made from skimmings, molasses and dunder. It is also generally asserted that rum can only be made from juice, or the residue of juice, which has gone through the process of sugar boiling; though what particular part of that process it is which constitutes the deciding influence is not known.
My experiments lead me to conclude that it is the heating of the juice with lime which is the essential part of that process.
I have already shown in the Bulletin for September, the part which I believe the treatment of the juice with lime to play in the production of the fruity acid in rum ; though it is probable that the nature of the influence is different to that which I suggested in article No. 2 of this series. Experiments which are not yet completed, seem to indicate that the fruity acid is an oxidation product of the essential oil ; at any rate observations extended over a considerable period, show that the fruity acid in samples of cane juice which have been treated with alkali and then sterilised increases considerably on standing exposed to sterilised air. This change whatever may be the nature of it, takes place in alkaline as well as in acid solutions, though in the former case the change is of course not discernible, until the liquid has been acidified. The influence of the lime in that case is antecedent, partial, and indirect, but lime also exercises a direct and complete influence on the aroma, in setting free the essential oil of Rum.
The way to prove this, when once this has been discovered, is very simple; since if we take and extract by a suitable solvent for this oil, and chloroform answers this purpose the best, a sample of cane juice which has been divided into three parts, of which No. I, is extracted without any previous treatment, No. II, after it has been boiled without lime, and No. III after it has been boiled with lime; and if we find on the evaporation of the chloroform solutions from these portions, that fractions I and II, leave no aromatic residue, but that fraction III does, we have proof positive that the action of lime, or caustic alkali, is necessary for the production of these aromatic residues. This has been the method employed in these investigations.
The considerations which led to this discovery were :—
1. The use of alkali in enabling No. 18 yeast to produce its characteristic aroma.
2. The failure to extract any aromatic oils from untreated cane juice, or from rum made entirely from such cane juice.
3. The discovery of this essential oil in Rum made in the ordinary manner.
4. That the treatment of the juice with lime, is the only strictly chemical treatment, to which the juice on the majority of estates in this country is subjected.
The odour of this essential oil may be best described by calling it the essential oil of Rum, since it has a peculiar and indescribable smell of rum, and, without doubt, it is to the presence in it of this oil, that rum owes its characteristic aroma; an aroma which is so utterly different to that of any other spirit. It is of an oily nature, and a small quantity will remain for a considerable time in a glass exposed to the air without entirely losing its strength; and I should say therefore that it is to this property that rum owes its oiliness, “body” and permanency of aroma which distinguishes genuine rum from any artificial imitation. After having been set free by caustic alkali from its combination in a natural state, it does not appear to be further acted upon by alkalis or dilute H2SO4 ; but strong H2SO4 seems to dissolve it with formation of a faint pink colour, which may be due to carbonization. It is readily soluble in chloroform, alcohol, and water. If a portion of cane juice, containing this oil, be filtered absolutely brilliant, the quantity of oil which can be extracted does not appear to have been lessened by filtration. Of course, this does not dispose of the objection that it might be found floating at the top, or forming a layer at the bottom of the liquid, in the form of an oil insoluble and lighter or heavier than the liquid with which it was associated; but, on carefully scrutinizing the filtrate, I have not been able to observe any such phenomenon. It is true that if we take a sample of cane juice, heat it with lime, and then let it settle and absorb air in some deep and narrow vessel, a Nessler’s reagent glass or test-tube for instance, that a band of darker colour may be observed at the top on cooling, but this is due to oxidation, probably of glucose compounds, and, on shaking the tube, the dark brown band at the top mixes easily with the lower portion of the liquid. It is also true that a waxy film may be observed floating on the top of such liquor, but it possesses merely a wax-like aroma, and is probably palmitic acid. That there is a large amount of wax in cane juice has been known for a long lime, and palmitic acid has been found in rum. (Mulder. Jahresbericht für Chemie, 1858, p. 302.) It would appear, therefore, that in the process of manufacture a considerable portion of this aromatic oil will be lost, since being soluble in water it cannot be separated by the preliminary clarification process in the cyphons, but must pass from thence into the coppers, where a good deal of it will be dissipated in the form of vapour. To a certain extent no doubt as the liquor gets concentrated by ebullition, and the essential oil has less water for its solution, and as the density of the liquor increases, it may rise to the top in an insoluble form, in which it can be skimmed off. But on these points further investigation is necessary.
Its boiling point appears to be comparatively high. It comes over from the still in the later runnings and can be extracted from dunder in considerable quantity, and I have succeeded in extracting in small quantity from molasses. The fact of its boiling point being high and that it can be extracted from dunder, shows us at once that we do not obtain by our distillation process, so much of this oil in our rum as we might do. It seems also to indicate that the use of patent continuous stills with high rectifying columns, would certainly be attended with a minimisation of the aroma of the rum in so far as this was due to the essential oil; and it suggests to us as to whether or not this may not be the reason as to why a low still head produces a better rum than a very high one——this seeming to be a pretty general opinion among Planters. If this should prove to be the case, it would seem as if the addition of an extra retort filled with cane juice, which had been properly treated for the production of the essential oil, through which the alcoholic vapours would have to pass, would be a distinct advantage : though certainly care would have to be taken that the lime was not present in excess, as this would lead to a decomposing of the “fruit ethers” which might come over with the alcoholic vapours. It is interesting to note that something of this sort appears to be done in some parts of the colony though instead of lime, common salt appears to be used.
What was said in a previous paper concerning the possibility of the existence of different “fruity acids” in different canes and in different soils, applies equally to this oil of rum. But here again a chemical examination of juice from different sources alone can decide. I am inclined to think however, after having examined rums from different estates and from different parishes, that, with the exception perhaps of the “new leather” smell, which may be caused by the presence in the rum of some single substance possessing that aroma, and which is not found in other rums, that variation is more to be sought for in the quantity in which this oil is present than that different soils each possess a characteristically different essential oil. Turning to the results obtained in the examination of samples of juice taken at different times from the mill from the same estate, I have been surprised by the marked differences exhibited in the quantity of this oil present. In very many cases it does not appear to be present at all. So far the samples have been taken at random from the mill without any attention having been paid as to the soil – on which the canes yielding such juice were grown, or as to the nature of the cane whether plant or rattoon. But having once established these two important facts, viz., the necessity of heating with temper lime or fixed caustic alkali to set this essential oil free, and that the amount in which it is present varies so widely, we are now in a position to carry on a systematic examination of cane juice from different soils with a view to ascertain what is the determining cause of such variations ; while it is evident that until such variations had been proved to occur, that any investigation in this direction could not be undertaken. It may be a matter of surprise to some, that if such variations do occur, that a greater variation in the quality of the rum from the same estate is not noticed. But the truth is that from the nature of our manufacturing process such changes in the character of the juice and the rum manufactured from it, must to a great extent remain hidden from us. In the parish of Westmoreland at any rate, it is the custom to mix and grind the canes from different fields simultaneously, so that the liquor, which finds its way to the still-house, is in no sense a representative sample of any particular cane piece, and the same may be said of the molasses. Coming to the still house, we not only use the liquor of a previous setting, i.e., the dunder, but many estates preserve their dunder from crop to crop; and in so far as concerns the distillation, not only is a double still frequently used, but also a retort containing a charge from the previous distillation, while, finally, the liquor is stored in butts of large capacity before racking into the puncheons. Yet, in spite of all this, anyone who watches the rum carefully, cannot fail to notice variations, and it is notorious that a difference in the quality of the shipments, from the same estate, is often commented on in England.
THE ORGANIC BASES.
But the treatment of the juice with lime also finds expresssion in the aroma and taste of the rum, in a distinctly unfavourable manner. It is owing to this that the organic bases or bodies of an alkaloidal nature are set free and are found in that state in the rum. A sample of rum shaken up with chloroform left behind, in addition to the essential oil, a vile smelling residue, the smell of which completely hid that of the essential oil, and resembled almost exactly the smell and taste of the dark brown liquid which may be found in the stem of a foul tobacco pipe. There seems very little doubt that this must be due to the presence in the rum of some organic base of the pyridene group, and last crop I succeeded in producing the same smell in cane juice by heating it with excess of caustic lime. During this crop, curious to relate, I have not been able to detect it in the samples of cane juice which I have up to now examined, but it must be understood that no systematic search for it has been carried out. But I have never yet failed to detect the presence in large quantity of some kind of bases or alkaloidal bodies in cane juice which I have heated with excess of lime, and I have so far never yet failed to detect them in greater or less quantity in Jamaica rum. The influence of these bases on the rum is to make it dry and harsh, and I should say, judging from personal experience, that the action of rum on the human system containing these bases in any appreciable quantity is not beneficial There seems very little reason to doubt, that it is the presence of these bodies which renders new rum almost undrinkable. Anyone who has had an opportunity of smelling these vile compounds, when isolated from the more agreeable smelling constituents of rum, could hardly be in doubt as to the desirability of doing away with them if possible. But can this be done? There seems to me no doubt that it can.
If cane juice be tempered with gradually increasing quantities of alkali, starting say from an amount sufficient to produce neutrality, it will be found that these vile smelling substances can only be extracted when a considerable excess of alkali has been used, and that by using moderate quantities of alkali it is quite possible to set free the essential oil of rum, without setting free these vile smelling bases. It will thus be seen that the tempering of the liquor is of great importance, not only to the process of sugar boiling but also to the manufacture of rum. As a result of my experiments in this direction I am able to say, that whenever the liquor is tempered until it becomes of a deep fiery orange to a ruby red colour, that there is danger of these organic buses having been set free.
It would appear then that a moderation in tempering is an essential point in the preparation of a good drinking rum, inasmuch as it would tend to furnish a rum of a mild, soft character, devoid of that dryness and harshness due to the presence of these organic bases, and thus fitter for immediate consumption : though whether this would equally hold good in the case of German Rum, in which I believe as much “flavour” as possible is desired, may remain a disputed point. On the other hand, on estates with heavy badly drained clay soils, rather heavy liming is, I believe, found to be necessary, which probably stands in intimate connection with the destruction of the glucose, which would be found in larger quantity in canes grown on lands where all the conditions necessary to the thorough ripening of the cane were not present. Besides the effect on the flavour, these organic bases have a very disastrous effect on the fermentation. As the result of a large number of fermentation experiments, performed with No. 18 yeast, I have found this to be invariably the case, and this enables us to make a shrewd guess as to what would be the effect of such bases on the human system.
The yeast is in fact poisoned. The amount of lime then used in the tempering of the liquor may not only have an effect on the flavour of the rum, but also on the progress of the fermentation, and it does not seem at all improbable that some of the disasters of the Still House may be due to the treatment of the juice in the Boiling House. I hope, however, at some future time to be able to furnish some more definite and interesting information on this point.
Lime therefore exercises a potent influence both for good and evil. Thus indirectly, it sets free the fruity acid. While directly it sets free—
(1). The essential oil of rum.
(2). Various kinds of organic bases or alkaloidal bodies.
As regards the nature of the chemical reactions which take place, it seems to me that we can pretty well infer it from the nature of the reagent used, and of the substances set free. The only reactions which can take place as far as I am aware are those of hydration and interchange of lime for the volatile organic bases or alkaloids combined with acids. In tempering cane juice, which for this experiment may be most conveniently performed by an easily soluble fixed alkali, such as caustic soda, it will be found, if the alkali, be used with caution, that a gradual disappearance of the fixed alkali takes place as the heating progresses. In this respect it is instructive to compare the behaviour of the two indicators phenol-phthalein and litmus; the former showing us the gradual disappearance of the fixed alkali, while from the behaviour of the latter, we should imagine that the fixed alkali was still present in excess. That this is not so can also be proved by distilling the liquor, when the distillate will be found to behave exactly the same towards the two indicators, blueing litmus and having no effect on phenolphthalein-—due of course to the setting free of volatile organic bases, which have no effect on phenol-phthalein: it need hardly be said that this is not due to the liberation of ammonia. But whatever conclusions may be ultimately arrived at as to the exact nature of the chemical changes, and as to the nature of the products of such decompositions, it will not I think be denied by anyone who carefully goes into the matter that the tempering of the liquor in the Boiling House has a very decided influence on the aroma of the Rum. No one who examines the residues which result from the chloroform and ether extractions of cane juice, before and after the treatment with lime, can have a doubt of this while an examination of almost any rum, which has been manufactured in this country, using the aforementioned solvents, will enable him to discover the same smelling substances which can be extracted from treated cane juice. It is not of course contended that the whole of the aroma of Rum is due to this cause, but I do not hesitate to say that directly and indirectly a very considerable part of it is. And moreover in breaking fresh ground, as I believe I am in these researches, it was considered more important to attempt to localise the chief points of interest and to ascertain the actual facts of the case, rather than to dwell on the theoretical side of the question, which becomes of importance only in relation to facts. These researches too were commenced in the first instance with one object in view, viz to elucidate the nature of the action of N o. 18 yeast in producing its characteristic aroma. I very soon found that I was working in the dark on this very interesting question, and it is only now on having learnt the nature and the result of the action of temper lime on cane juice, that I am encouraged to believe that I have found the solution of this previously difficult problem. But of this more anon. It is obvious however that even the practical side of the question is far from being exhausted, and an ever broadening field of investigation is opened up. If one may be allowed to theorize a little, there seems sufficient grounds for concluding, from the results which I have up to now attained, that though the aroma of rum is in the first instance derived from the soil, that this influence is chiefly potential not actual; that it is latent, dormant, and only brought into existence during the process of manufacture. If this should prove to be the case, it would seem to hold out a hope that much may be done to improve our rums both for the home trade in England and for export to Germany ; and to imply a rebuke against murmuring with folded hands—“it’s the soil that does it.”
“SELECTED” YEASTS AND GENERAL CONSIDERATIONS.
By Percival H. Greg.
In considering the question of the adaptability of the “selected yeast” system to our Jamaica Still Houses, a system which has found such favourable acceptance in other countries, we must first of all consider what are the requirements of the manufacturer of rum and the conditions under which he labours. I may say without fear of contradiction that wherever by previous systematic experiment and by a careful study of the circumstances of the case, the right type or variety of yeast has been found, that the application of the system has been an unqualified success : and that the right type or variety of yeast exists and may be found with the expenditure of a reasonable amount of energy is proved by the ever increasing numbers of Breweries and Distilleries working with selected yeast. [This is very big!]
It would however be a mistake to assume that this has been accomplished in a day, or that the principles when first enunciated by Emil Christian Hansen were favourably received. Far from it! The advocates of the system had to encounter indifference, ridicule, and active opposition. It would however be beyond the scope of this paper to enlarge on these points, and would indeed be superfluous, since now that an English translation of Hansen’s “Untersuchungen aus der Praxis der Gärungsindustrie” has been issued, those interested in the matter may drink in knowledge from the fountain-head. It is sufficient to say that the first success on a commercial scale was obtained by Hansen in the Old Carlsberg Brewery in the year 1883 or 12 years ago, and that since then the application of the system has not only grown steadily in the department in which it was first introduced, but the principles upon which it is based have found a successful application in many other directions. It seems natural therefore to question whether what has been so successful elsewhere may not succeed in Jamaica. At the first glance it seems evident that there is one very great difference to be distinguished as regards the application of selected yeasts between the manufacture of rum from pure cane juice and the manufacture of beer or spirit from cereals. In the case of beer, the “wash” or “wort” as it is called, i.e., that infusion of malt and hops which on being fermented yields beer, is after a lengthened period of boiling in the “wort” copper, and subsequent cooling on the refrigerators or “coolers”, brought into the fermenting tuns in a more or less sterile condition. Of course experiments have shown that a certain amount of aereal contamination invariably takes place, but in this case the germs are in a dessicated state, and it is pretty certain that by far the greater number of them find such a strongly hopped medium as ordinary beer-wort unsuitable for their growth in any considerable measure, at any rate during the primary fermentation ; and of course great care is taken to “pitch” the wort, i.e. add yeast to it, immediately that oxygen in sufficient quantity has been absorbed, and a favourable temperature for fermentation has been reached. If then the wort be pitched with a selected and suitable type of pure yeast in sufficient quantity, there is very little to fear from the competition of foreign yeasts, i.e. yeasts other than the type intentionally employed.
In well-conducted Distilleries the case is practically the same, though here an intentional lactic acid fermentation in the yeast-mash is caused, in order that the antiseptic properties of the lactic acid which is formed, may protect the yeast from hard usage at the hands of other bacteria. After a sufficient amount of lactic acid has been produced however the lactic acid bacteria are killed, or at least rendered hors de combat by warming the yeast-mash up to 60° Réaumur. The mash or “wash” which serves for fermentation is also rendered sterile for all practical purposes by conducting the “saccharification process” at a comparatively high temperature, so that here again the desired type of yeast has the field practically to itself. But in the case of the fermentation of sugar cane juice the case is somewhat different. If we wish to give ourselves an idea of the results to be expected in the application of selected types of yeast in the fermentation of fresh cane juice, we must examine the results obtained in the application of this system in the preparation of wine, i.e., in the fermentation of the juice of the grape.
Grape juice and cane juice possess the property in common of entering into fermentations “spontaneously”—if left to themselves they begin to ferment without the addition of yeast.
Pasteur proved many years ago, see “Studies on Fermentation,” an English edition of which is published by Macmillan & Co. London, that this was caused by the yeast cells which are found adhering to the skin of the grape, and on the crushing or squeezing of the grape come into immediate contact with the juice, and cause it to ferment. Recent researches have shown that the same process obtains by the cane. We can easily see therefore, that in order to obtain successful results, we must either get rid of the yeasts naturally present in the cane juice, or we must be certain in advance that in inoculating the juice with such and such a yeast, that we do it with that particular yeast which is absolutely adapted for the work it has to perform and which will find the medium in which it has to ferment in all respects a suitable one. If not, then we run the risk, that the struggle for existence which invariably ensues will ultimately end in “the survival of the fittest” which may not perhaps give the kind of fermentation we desire. The following translation of an abstract of a lecture by Julius Wortmann, delivered before the 13th Congress of the Wine Trade in Mainz may prove interesting. This article is taken from the “Centralblatt für Bakteriologie und Parasitenkunde,” and the title of this article is (translated). “The practical results obtained up to the present time with pure yeasts, and the practical lessons to be learnt with regard to the selection and application of the same.”
“While the brewing and Spirit-Making industries have long ago made use on a practical scale, of the favorable results which were obtained in the Laboratory by Hansen’s system of fermentation, it is only lately that these magnificent results have attracted the attention of those engaged in the manufacture of wine. It is but a few years back that the first attempts were made, to ferment grape juice by selected yeasts from the Laboratory. Up to then no attention had been paid to the yeast in grape juice fermentation, which was simply brought about by means of impure races of yeast which were derived from the grapes to which they were fortuitously attached. Such a fermentation by means of these races of yeast of unknown source, along with the simultaneous growth of all manner of moulds and bacteria, it is now attempted to displace by an addition of pure cultivated yeast to the “Must” (grape juice) and to bring about in its place a purer fermentation of a previously known character. Further it is desired by this means to do away with the products of decomposition of the organisms originally present in the Must, which frequently exert a disagreeable influence on the taste and aroma, and by this means obtain a purer tasting more even charactered wine. After experiments in this direction, in the Laboratory of the Research Station at Geisenheim had led to favourable results, attempts were made on a practical scale with selected races of yeast, cultivated pure. Since as already mentioned the yeast which, was added to the “must” had to enter into competition with the organisms already present there, the pure yeast must be added (1) in sufficient quantity, (2) in active fermentation, and (3) when possible should be added to the “must” before it has begun to ferment. These conditions however, especially point 3, were in the latter end of the warm and dry autumn of 1893, very difficult to attain, because in consequence of the warm weather the yeasts attached to the grapes were very strongly developed, and thus swarmed in the expressed juice in large numbers, and at the same time the temperature being high, fermentation began quickly so that in by far the greater number of cases, the pure cultivated yeasts could only be added to the “must” when it was already in a state of fermentation. In order to insure that the pure cultivated yeasts were not added to the “must” in a too weak fermentating or even partially inert state, the Research Station did not give the same to the wine-makers in accurately measured quantities sufficient to ferment a fixed quantity of must, but every fermenter received the freshly cultivated yeast in a few litres (1 litre—1.37 pints) of previously sterilised must when this must had nearly finished fermenting. According to instructions given with this, he (the fermenter), had then to cause this yeast to multiply by gradually adding to it larger and larger quantities of must, until he had by this means obtained a quantity of yeast sufficient for the whole quantity of must to be fermented. This method was somewhat detailed for the fermenter, but he obtained by this means a sufficient quantity of yeast in active fermentation. In those cases in which it was possible by this method to bring the yeast into the must so that it obtained the upper hand from the commencement, favourable results were not wanting. Almost without exception the musts fermented by the pure cultivated yeasts were distinguished from the musts fermented in the ordinary manner, by a quicker and more intense fermentation and a more pronounced bouquet. . . . . .”
The italics here are my own. In regard to what Wortmann writes concerning the state of activity of the yeasts naturally present in the grape juice, we may I think logically conclude that the same will be the case in a greater degree in cane juice, owing to the high temperatures prevalent in the Tropics: in fact my own experiments in this direction confirm his views. Indeed it is very much a question in my mind, whether, taking these difficulties into account, the “game would be worth the candle.” This point however can only be decided by a course of experiments on a large scale. But the real fact of the matter is, that the total amount of fresh cane juice fermented in this country represents but a small fraction of the total amount of liquor which passes through the distillery. By far the greater bulk of the wash, as is well known, consists of skimmings, molasses, and dunder. But it is precisely for this reason that I consider that systematically selected types of yeast might be of such great service. In the first place, the “tempered” cane juice will, on the generality of estates, have been steamed in the Syphons, and though bearing in mind the resisting power of “spores” towards heat and the fact that the skimmings contain a large amount of wax which might enclose and protect some of the spores from the full effect of the heat employed, and further bearing in mind that we have no rigid experimental data to go upon in this particular instance, yet there seems very little doubt that their subsequent growth and development will have been retarded. As regards the molasses I have satisfied myself by microscopical examination, that in the generality of cases the microorganisms present consist principally of torulœ which possess but a slight fermentative power, and bacteria. These last might probably be very much diminished in number if proper attention was paid to keeping the sugar coolers and molasses hole clean. As regards the dunder this must evidently be sterile when taken from the still, and when not allowed to stand for any great length of time after having cooled down is probably but little altered.
It would seem then that as regards these liquids any particular cultivated yeast which we might desire to employ would not have to meet with any extraordinary opposition from other germs.
The subject naturally falls under two heads: First the effect of selected types of yeast in imparting (a) an increased regularity to the progress of the fermentations, (b) in giving the type of fermentation desired— slow or fast, and (c) in obtaining a greater yield of alcohol from a given weight of sugar; second, the effect of selected types of yeast on the flavour and aroma of the resulting products of fermentation, i.e., on the rum. In a previous paper (The Jamaica Yeasts Bulletin for August) I showed in some tables that other things being equal, the whole character and nature of the fermentation varied with the kind of yeast employed. We may therefore dismiss questions a, b and c as not requiring discussion but before doing so we must consider in how far the character of the fermentation produced by a given type of yeast may be modified by circumstances.
It is obvious, since yeast is a living organism, that unless the medium in which it has to live is suitable to it, it will not exercise its functions completely, i e., will not assimilate or feed, will not grow and consequently will not ferment in the same degree which it could if the conditions were entirely favorable. It must not therefore be hastily assumed, that by using pure cultivations of selected types of yeast that a mere mechanical regularity in the fermentation will be assured, without any attention being paid by the Distiller to the nature and composition of his washes. But it is evident that by working with a known quantity, as we should do when working with a selected type of yeast, which has been originally grown from one single cell, that we are able to study the conditions or sets of conditions which may be favorable or the reverse to the yeast in question in advance, and thus we shall know what to aim at and what to endeavour to avoid in our practical operations. Whereas in working with a mixture of yeasts, we are working with an unknown quantity. What we can say therefore is this, that other conditions being favorable, the fermentation and the yield will be much more regular by the use of one type or variety of yeast than by the use of a mixture of yeasts. Thus supposing for the sake of illustration, that we have as a mixture of yeasts No. 4 and No. 18, one of which ferments in 3-4 days and the other in 12 ; the one producing comparatively speaking but little aroma and the other producing a very strong and definite aroma, it is obvious that we want the one and only the one of these two yeasts according to the relation existing between the working capacity of our still house and boiling house and according to the character of the rum we require to produce. If in comparison to our boiling and grinding power our still house is large, we can afford to work with a slow fermentation, and there being no particular object then in emptying our vats quickly, we should no doubt be desirous of taking advantage of the aroma producing power of No. 18 yeast. Suppose that we do so desire, but that instead of working with 18 yeast cultivated pure, we work with a mixture of No. 18 and No 4. What guarantee have then that at some future time No. 4 may not obtain the upper hand over No. 18, and that instead of working with a slow aromatic fermentation, we may suddenly find our “washes” attenuating with startling and unwelcome rapidity, and the aroma produced by fermentation decreasing? Or suppose exactly the opposite case: that we have a large crop good boiling and grinding power, but a small still house. Here we are forced to work with a quick fermentation, and we desire a yeast giving a quick attenuation such as furnished by No. 4 yeast. Would it not then be very inconvenient to suddenly find our washes covered with a thick golden head, and lying apparently “lifeless” in the vats, with the density say at 20 arnaboldi? Yet I have seen and heard of such cases occurring in Jamaica still houses. What in one still house may be a perfectly normal and desirable fermentation may be quite abnormal and undesirable in another.
We now come to the consideration of the second division of the subject. What would be the effect of pure selected types of yeast on the quality of the rum? In other words, granting that there would be an increase in the regularity of the fermentation, and in the yield of rum from a given weight of sugar, and that we might produce a slow or quick fermentation at will. Would the quality of the rum deteriorate? Should we in fact by this method manufacture not rum but spirit? The answer to this question will depend to a certain extent as to whether we are working with the methods at present employed to make “quality” rum, or ”common or clean.”
In my second “Contribution to the Study of the Production of the Aroma in Rum” which appeared in the September Bulletin, I showed that my experiments led me to believe that a preliminary “souring” in the “trash” cistern was capable of playing an important part in the production of the aroma in rum. And the trash cistern has long been considered as a sine quâ non by practical men in the production of high flavoured rum. I have already stated that yeasts exist which are capable of influencing the aroma, and among the Jamaica yeasts which I have isolated there is one No. 18 yeast, which not only influences the aroma in a very marked degree, but it is also capable of inducing a normal alcoholic fermentation. It is an open question therefore whether an aromatic rum might not be produced by an absolutely pure yeast, without the intervention of a “trash” cistern at all. But nothing but rationally conducted experiments in the still house can decide this question. Even if this should be proved to be possible, it would not be sufficient; the whole question has to be considered, to a certain extent, from a commercial point of view. Experience goes to show that it is not only necessary to produce an aromatic rum to gain a good price, but to produce rums of different aromas. As far as I can understand, a particular kind of rum, which for a few seasons may command a high price, may afterwards lose the price even though the character and quality have not deteriorated. I believe “pineapple” rum and “new leather” rum are instances in point.
Therefore it behoves us to look at the matter in this wise. Can the selected yeasts system and the “trash” cistern be worked in conjunction? Does the one method render the other impossible? In other words we must consider the question with reference to all the means by which it is possible to produce aromatic rums. My own opinion is that taking into account the manner in which the duty is levied by the Excise, not only can this be accomplished, but that the introduction of a methodically selected yeast into a still house may be a great help to the efficacy of the working of the “trash” cistern. Quite otherwise might it be under another set of conditions : thus in Germany a longer period of fermentation than three days is not allowed.
What is the general opinion among practical men regarding the “trash” cistern? That it produces good rum but “slows” the fermentation! And the result of this? That good rum can only be produced on estates making small crops in comparison to the size or working capacity of the Still House. There is another consequence of the trash cistern, viz, that the yield is poor. This is so well known that it has come to be a generally accepted axiom, that quality and quantity cannot be produced simultaneously. This is certainly true under the existing state of things. It is possible that it will be true under any system of fermentation, but I doubt very much whether it need be true to the same extent under an improved system of fermentation, by means of selected yeasts working in conjunction with the trash cistern.
Why does the trash cistern slow the fermentation? I will answer it by another question. What are the changes which take place in the trash cistern? Says the practical man, “the liquor rots,” ” ripens” or, “decomposes” in it, which is very true—only rather vague? What does take place beyond a doubt is a production of acidity. But is not an acid medium favourable to yeast? It depends entirely on the nature and extent of the acidity produced. This is obvious. As I have already stated, in the production of spirit from “Cereals” lactic acid is intentionally produced in the yeast mash or “good,” because it has been found by experiment that lactic acid is an antiseptic and favours the yeast by suppressing by its presence the growth and development of other bacteria which are distinctly harmful to yeast. But though lactic acid may to a certain extent be produced in the trash cistern, acetic and butyric acids are also largely produced. That acids such as acetic and butyric exert an unfavorable influence upon the growth and development of yeast has long been known for a fact in Europe ; that the same obtains here I have myself proved in the Laboratory. Some extent of the damage done may be inferred from the following figures which are taken from Mœrckers Handbuch der Spiritusfabrikation (Handbook of Spirit Manufacture) p. 461. [table linked here]
It will be seen that up to 0.03 of 1% the amount of alcohol formed is slightly increased from the case in which no butyric acid is present, in this case the increase in the amount of alcohol amounts to [?? some fraction] of 1%, but when the butyric acid present amounts to something between 0.03 and 0.1 of 1% the production of alcohol is nil for practical purposes. It will be noticed that these figures refer only to the influence on the yeast of the life products of the bacteria, but there is a special influence exercised by the bacteria themselves irrespective of the various acids which they produce. It is not the place here to enter into a discussion on this point, it is mentioned only to show that the evil could not be remedied by neutralizing any excess of acidity which might be produced. Taking into account then the influence exercised by bacteria and their products, it is easy to understand that if we depend upon the trash cistern to furnish us with a supply of vigorous yeast to produce a fairly rapid and satisfactory attenuation, we shall be disappointed. The yeast will have been already weakened in its fermentative and reproductive capacities, and consequently the fermentation will be slow, that is to say the rate of attenuation and the yield of alcohol will be unsatisfactory. I am fully aware that trash cisterns are worked with the result of giving a fair average attenuation and yield viewed from the Jamaica standpoint, but then they are worked more on the lines of a “mother” cistern, i.e. are prevented from going radically sour by being always kept alive by periodical doses of molasses and dunder; but in these cases I have not seen real high priced rum produced. But when we understand that a great deal of the unsatisfactory nature of the fermentation may be referred back to the weakened yeast, we are inclined to enquire whether this cannot be remedied. Supposing that we cannot when working with, the trash cistern have a fermentation which would compare in rapidity of attenuation and percentage yield of alcohol with an ideally pure alcohol fermentation, what amount of slowness is necessary for the production of good rum? Is it absolutely necessary that the fermentation should last 2 or 3 weeks, or could it not be reduced, let us say, not to be too ambitious, down to 7-10 days without injuring the quality of the rum? And supposing that the yield must suffer what is the limit of the amount of quantity to be sacrificed to quality? In other words, may we not to a great extent be following the example of the ancient Chinese in the fable, in that we are burning down our house to roast our pig, when the same thing could be done just as effectively with the expenditure of a less amount of valuable material!
It must be remembered that this acid fermentation in order to produce an aromatic spirit is not peculiar to Jamaica. The bouquet “Whiskies” which have no doubt played a part in diminishing the consumption of Jamaica Rum are manufactured very much on the same lines. A certain portion of the “mash” mixed with dunder is allowed to undergo a 20 hour’s “souring” but the yeast is grown separately in a specially favorable liquid, before being required to ferment the “soured” mash. The whole thing is worked on a perfectly intelligible and rational theory, which has been evolved from experiment. Having produced a giving amount of acidity it is desired to stop it, and to produce alcohol. But how can this be done when the mash contains a large amount of volatile acids and bacteria which are distinctly harmful to yeasts? It has been found by experiment that the stronger and better nourished a yeast cell is the better is it able to resist a disease. Therefore by growing the yeast first in a separate favorable liquid and employing it when its development is approaching its maximum, we give it the best possible chance of overcoming the bacteria. We are also able by this means to introduce the yeast in large numbers into the mash because having been kept separate from the bacteria in its reproductive stages, it has been able to reproduce itself fully.
Therefore by introducing the yeast in sufficient quantity and in a high state of physical efficiency even into a mash which is swarming with bacteria after having undergone a 20 hours “souring,” it has been found possible to induce a very fairly rapid and satisfactory alcoholic attenuation.
The average attenuation in one of the biggest Distilleries in America where these so called sour mash whiskies are produced is: period of fermentation 4 days and attenuation 19° Balling down to 4°, the dunder used having a density of 4° to start with. This in Jamaica figures would be approximately 25.27—5.32=19.95 degrees attenuation. Now it suggests itself that the same principle might with advantage be applied here, that is to say one vat or ground cistern might be started in good fermentation by means of pure fresh cane juice, and as fast as it was “cut” out into other cisterns, filled up with a suitable mixture of pure molasses and dunder, so as to keep it always alive meanwhile the souring process might go on in the trash cistern. This is very much the modus operandi that would be followed if a yeast propagating apparatus was used, the general method of working it being, to take the yeast or the fermenting liquor from the apparatus and set up a vat with it, the liquor of which should be as pure as practical considerations will allow, and then to “cut” from this vat into other vats. When this No. 1 vat as I may call it gets exhausted or impure, it is cleaned out with scrupulous care and a fresh supply of yeast from the propagating apparatus is put in and so on. As to the advantages which we possess in working with a yeast producing a fermentation of a preciously known character in comparison to working with mixtures of unknown yeasts, present in unknown and varying proportions, as would be the case if we started our No. 1 vat with cane juice, instead of yeast from the apparatus, I have enlarged on in the earlier portion of this paper, and I can only add that the surest way to attain perfection in a manufacture which is admittedly subjected to unaccountable and unlocked for changes, lies, in the elimination of the uncertainties, unknown quantities and potential causes of variation as far as lies in own power. It might therefore be possible by the selected yeast system working in conjunction with the “trash” cistern or working by itself not only to obtain a greater control over the progress of our fermentations to make “quality” rum at a less sacrifice of time and material, but to alter the character of the resulting rum to a certain extent at will. Thus further search might reveal the presence of other aroma producing germs besides No. 18. We might therefore in the future be able to produce an aromatic rum by a pure fermentation of 18 yeast or some other aromatic germ without the trash cistern. Again we might set to and manufacture an aromatic rum by means of No. 18 yeast working in conjunction with the trash cistern, or again a different charactered rum by any of yeasts No. 1, 4, 5,7, 8, 14, 17 or 19 working in conjunction with the trash cistern according to circumstances. So much of the case where it is desired to produce an aromatic rum. With regard to the production of common clean rum it would seem as if the introduction of a selected yeast giving a rapid attenuation and big yield could not fail to be of great service. It might prove of great value in producing a purer and more even charactered rum containing less by-products which might be fit for speedier or direct consumption and by this means the consumption might be stimulated.
As to this and all other questions raised in this paper I would impress up on those whom these matters may concern, that mere theorising is of no use at all. Laboratory experiments are indispensable in showing whether there is an a priori possibility of an improvement being effected or a reform carried out in these matters, but the deciding word can only be spoken after systematic experiments in the still house itself have shown whether the reform is practically possible. At the same time I am of opinion that much may be done in Jamaica still houses, and this is a very important point in these days, at a very much smaller expenditure of money than would be required to effect an equal improvement in our sugar manufacture. In the old days when the sugar crops were smaller in proportion to the working capacity of the still houses, “spontaneous” fermentation could be trusted to do the work, but now-a-days, owing to the sugar crops having increased, a great many estates habitually run their liquors before they are “dead” and the consequence is an inferior product all round is put on the market ; and the consumption of rum, though certainly not exclusively owing to this cause, has decreased. Greater demands therefore are made on the fermentating agent, i.e, on the yeast, and therefore greater attention and care must be bestowed on the yeast question than has hitherto been bestowed.