G. Saussine—The Chemistry of Rhum, 1899

Follow along: IG @birectifier

SAUSSINE (G.). — La chimie du rhum. Bull. Agr. Martinique (1) I, 101-109, 1899.

This citation came from Kervegant’s L’Industrie Rhummiere A La Martinique, 1933, and has been newly digitized to be translated here. The paper spends most of its time describing the work of Percival Greg of Jamaica who was among the first to identify fission yeast as important to rum production. The paper also looks at the mystery of high value aroma formation in rum production and what was known at the time. The most unique piece of information regards the trash cisterns used in Jamaica with an admission that fruit peels were commonly added, but it is hard to say if that relates to Jackfruit use at Hampden or use of fruits at DDL for their high ester rum. It is also posited that much high value aroma in Jamaica rum comes from the specific use of the trash cistern and is possibly their source of what we would now identify as damascenone. All the information is presented with the vibe that the ideas were fair game for implementing in Martinique. Possibly just a few short years later, it would be known that fission yeasts were part of heavy rum production in Martinique.

The Chemistry of Rhum
(Review of foreign works)

The chemistry of Rum is without a doubt one of the most difficult questions to tackle in a local publication; there are such commercial interests attached to it that many people, unacquainted with scientific discussions, are disposed to regard the development of such a subject as a kind of profanation. However, this question has been the subject in France and abroad of important works which it is useful to spread among the producers of the Colony. An author who has himself practical experience in the manufacture of Rum published a few months ago in a local newspaper a compendium of experiments made in Jamaica by Mr. Percival H. Greg, and I will only follow his example by taking up the same question here with the developments it involves.

Another difficulty encountered in dealing with such subjects is that they require from the reader certain fundamental knowledge, unfortunately not very widespread, on the chemistry of alcohols and physiology of yeasts. It is necessary to come back to the general phenomena already acquired to clearly show what recent studies bring that is new and particularly interesting for our local industry.

The work of Mr. Percival H. Greg brings a certain number of new facts: this work comprises two parts: the separation of the yeasts which one meets in the vats of the rhummeries and a study of the conditions in which certain aromatic products appear during manufacture.

I. Yeast selection

The separation of the different breeds of yeast which apply to the same industry, for example brewing or wine-making, is a rather long operation into the detail of which I cannot enter. In the absence of very clear external characteristics, the experimenter who has isolated a certain number of yeasts which he believes to be distinct, must try them on a small scale to see if they behave differently in a given nutrient medium. By changing the nature of this medium, the temperature or the acidity, we observe variations, either in the progress of the fermentation, or in degree or aroma of the product obtained.

These results are well known for beer; they have been applied to vinification, where the use of selected yeasts has made it possible to enhance the bouquet of ordinary wines (1). They have also been applied in the distillery and this is what has been tried on various sides for sugar cane yeasts.

(1) Jacquemin: amélioration des vins par les levures sélectionnées (1898).

One of the first, M. Marcano, described a yeast found on cane in Venezuela (2) presenting some interesting particularities: it is smaller than the alcoholic yeasts of temperate countries; the spores, on a very nutritious medium, reproduce a mycelial form. It is probably this yeast that M. Delafond mentions in the distilleries of Venezuela (3).

(2) Marcano: comptes-rendus de l’Académie des sciences (1887).
(3) Delafond: Bulletin de l’Association des Chimistes (Septembre 1898).

M. P. H. Greg has undertaken a comprehensive study of rum yeast. Their description is reported in German (4) and English reviews which I do not have at my disposal; but the author gives some indications in the Bulletin of Botanical Department, Jamaica (August 1895). Among the various races which he isolated, there is one which particularly interested him and which he designates by the number 18.

(4) Centralblatt fur Bacter und Parasitenkunde (1896).

While the others were bottom fermenters, number 18 is a top fermenters: it formed a yellow creamy cap on the surface of the liquid; it ferments very slowly but ultimately gives a very high alcohol yield. Compared with some other rum yeasts, it gave the following figures:

We use Baumé degrees more often at home. To convert degrees Brix to Baumé, simply divide the given numbers by the factor 2.77. Thus 21 Brix = 7.5 Baumé; 5 Brix 1.8 Baumé.

It is customary in rhummeries to assess the yield by comparing the volume of commercial alcohol obtained with the volume of syrup used. This report has no value for chemical manufacturing control (1); this is the volume of absolute alcohol, contained in 100 liters of the grappe in fermentation.

(1) It would be necessary to compare this volume, or the corresponding weight of alcohol, with the actual weight of sugar introduced.

Yeast No. 18 presents the deepest attenuation with a very high yield, but the slowness of fermentation seems to be a difficulty for its industrial use. On the other hand, it has a quality that makes it valuable, which is to give, all other things being equal, a very aromatic product. The other yeasts, on the contrary, give an alcohol without aroma, or at least of a very inferior aroma.

In opposition to No. 18 is No. 4, bottom fermenting yeast with rapid fermentation, whose alcohol yield is a little higher; but it gives no aroma.

From the point of view of alcohol production alone, if we observe that for a 2000 liter tank, a difference in yield of 1% corresponds to nearly 40 liters of commercial alcohol, we see the interest that attaches to yeast choice.

The author points out that the aroma of no. 18 does not appear during the course of fermentation, but only afterwards, when the fermented liquid is left to itself for a day or a day and a half; yet it only appears under certain conditions which remain to be specified. Therefore, if we want to obtain a product that is always identical, we must ensure that we are always dealing with the same yeast or at least that it is predominant in the fermenting vats. It becomes necessary to cultivate that yeast at the factory in a special apparatus (1. Several models of these apparatus are in use in distilleries; I will cite, for example, among the French apparatus which have been described, the model of M. Fernbach (3).

(2) Jacquemin: Les levères sélectionnées en distillerie (1896).
(3) Kayser: les levures. Encyclopédie Leauté,

M. Landes, in 1894, took out a patent for an apparatus intended for the same purpose. However, most distillers turn to special houses for regular supplies.

II. Aroma Production

Chemical analysis has so far apparently been powerless to detect in rum what constitutes its special aroma. M. Percival H. Greg, comparing the very detailed analyzes given in the German works of Dr. E.-K. Windisch and Dr. E. Sell with the prices obtained by the products submitted for analysis, concludes that the price differences cannot be attributed to any category of the substances measured. But there still remain in the chemical study of rum many obscure points, the solution of which presents the most serious difficulties.

Thus, by fermenting yeast no. 18 in different media, refined sugar, dextrose, cane molasses, cane juice, brought to a sufficient dilution with water and added with the necessary nutritive salts, one does not obtain the characteristic rum aroma. If the same sugary materials are diluted with the dunder which results from a previous fermentation of the same materials, the aroma appears sometimes but not always. This aroma is therefore not due solely to the yeast. To specify the conditions under which it appears, Mr. Percival Greg reports some experiments of which here is the clearest.

A certain volume of cane juice is divided into three batches brought to the same dilution:
No. 1 is alkalized with caustic lime;
No. 2, serves as a control, does not receive any treatment;
No. 3 does not receive lime, but it is weakly acidulated with sulfuric acid.

To protect from any foreign fermentation it was necessary to sterilize the liquids by a long boiling: no. 1 is subjected to boiling after liming, to imitate industrial treatment; no. 3 is boiled before acid addition. But, as yeast does not work in an alkaline medium, once sterilized No. 1 is added more sulfuric acid. In these three media, the same weight of yeast No. 18 was inoculated.

When the fermentation is finished, the liquid is left to rest for 20 hours: no. 1 has developed the desired aroma, but quite weakly. The other yeasts subjected to the same control did not give aroma or led to a different aroma.

Other experiments, which are not presented with the character of sharpness of the preceding, seem to indicate that when dunder is employed, with molasses as well as with cane juice, only the dunder coming from operations in which No. 18 has worked, communicates to the product of fermentation the particular character of rum. By associating this dunder with sugar cane previously limed, then sterilized and acidulated, or simply with syrup, one obtains a more accentuated aroma than in the case of No. 1 of the experience reported above.

The author does not seem to have focused on the distinction that is made here between the fermentation of fresh cane juice and that of syrup. No doubt this distinction was of no interest to the industrialist of Jamaica, if we are to believe the descriptions that we have of the processes in use in that country; but it seems rather that he did not find a sufficient theoretical reason to stop there, seeing in molasses and dunder only a residue of cane juice and concluding that the aromatic products contained in these residues must exist already in the cane juice. It might be objected that some of these products may have been introduced by the action of yeast, but this yeast is the same in all experiments, is no. 18.

This would be the place to recall how certain industrialists in Jamaica practice fermentation. This operation is summarily indicated in the report of M. Landes (1).

(1) Supplément au Bulletin Agricole, page 80. [I found and translated the brief supplement and attached it at the end of this document.]

Near the sugar refinery there is a pit, trash cistern, or dirty cistern, filled with bagasse where the scum from defecation and residues from filter presses are collected. Fruit peelings are sometimes added. Various ferments which live on the cane or on the fruits develop abundantly on such a nutrient medium: alkalines quickly neutralized the products of fermentation that soon develop a frankly acid reaction. It is a very energetic yeast but very variable in its composition. Experience has taught manufacturers that the use of this yeast in fermentation tanks favors production in the alcohol of an oily substance to which they attach great importance.

[Fruit additions to cane trash cisterns may have a few possible origins. The first was an understanding of fermentation adapted from Liebig where decomposition of nitrogenous material started fermentation as opposed to Pasteur’s later yeast-centered approach. Something that decomposed faster than the cane could be a beneficial starter. The second origin was a fascination with the way fruit leaves such as peaches contributed aroma during fermentation. Wray was exploring peaches and even pineapple skins. Exposing the cane stalk to fermentation was an extension of that. What ever operated successfully such as a peach of pineapple could spread its success to the cane. Some of this is now cellulosic fermentation but it is barely understood in a rigorous way for this context.]

It is important to know if it is indeed to this substance that we must attribute the special bouquet of rum and how the reactions which take place in the pits make it appear. If we refer to the opinion of the German authors, this bouquet would be mainly due to two substances that the analyzes do not give:

1. A fruit-smelling, high-molecular-weight aromatic acid that Herzfeld acknowledged existed but did not
2. An essential oil that has never been well studied but is believed to belong to the terpene group, like most plant essences, such as lemon, mango, pepper, etc. M. Percival Greg has brought to the study of these two substances some new experiences, not from the point of view of their determination, but from the point of view of the conditions in which they appear.

Aromatic acid.—Its smell is quite difficult to define: it recalls at the same time grape must, acetic ether and lemon zest: to abbreviate the author simply designates it under the qualifier of fruity. He found it in aged grappes put in reserves at the end of a campaign, with a view to preserving the ferment. They didn’t seem to offer any special aroma, but rather the smell of vinegar. The acetic acid seems to have gradually diminished and the smell of rum has appeared. Petroleum ether made it possible to extract a very fragrant product, the odor of which disappears by adding caustic soda and returns by adding sulfuric acid. It is therefore an acid.

[This paragraph may be referring to the “bauer oil” part of the rum oil puzzle as opposed to the rose ketone, damascenone. I’m not quite sure if aged grappes refers to accumulated tails of distillates, but that is my guess.]

Did this acid exist originally in the vesou? We have seen that the treatment with sulfuric acid does not produce an aroma which must be limed, cooked and acidified; in the juice thus treated, petroleum ether makes it possible to extract the same aromatic product. This acid therefore already existed in the cane, perhaps in the form of a solid ether which the lime would have the effect of saponifying. But Mr. Percival Greg returning to these experiments in an article published a few months later (1) thought it necessary to give another interpretation.

(1) Bulletin of Botanical Department Jamaica. (January 1896.)

Essential oil.— The acid in question would be the product of slow oxidation of an aromatic essence of cane. This oxidation would take place during manufacture under the circumstances which have been indicated, namely when the juice was stirred and cooked, then subjected to the slow fermentation of yeast No. 18. This essence can be isolated by means of chloroform, but only after the action of lime.

A sample of cane juice was divided into three batches: the first was stirred with chloroform without any preliminary treatment, the second was subjected to boiling without the addition of lime: the third was previously limed and cooked. It was only with the third that the chloroform extraction left an aromatic product absolutely reminiscent of the smell of rum: it is an oily liquid, not very volatile, which does not appear to deteriorate in the air and does not lose its aroma in the presence of alkalis or dilute acids. Concentrated sulfuric acid dissolves it with a brown coloration. This essence is soluble in alcohol, a little in water; it seems to be carried along in part with the steam in the cooking apparatus. During distillations, it passes into the tail products and is found in large part in the dunder. This explains the experiments reported above, namely that the aroma is developed not only by the addition of lime to the cane juice, but also by the use of dunder resulting from an identical operation.

If the aromatic acid mentioned above is only an oxidation product of this essence, is the action of yeast no. 18 due to a special physiological property, or simply to the slowness of the fermentation which allows contact with the oxygen in the air for a longer time? We see that the question is far from being elucidated; but the experiments of the English chemist trace the directions in which research must be done and in which he himself had to advance since the publication of this work which dates from 1896.

Let us add that the aromatic acid in question here would have been identified according to certain authors with pelargonic acid which is in fact a product of oxidation of an essence; but the identity of this essence with that which is indicated in the cane is far from being demonstrated. Finally, according to Mr. P. H. Greg, this essence of cane should not be confused with the waxy matter which has been known for a long time and in which palmitic acid has been found; but it is mixed with this wax and other substances of which we have to speak.

Disadvantage of excess lime.—We still owe to Mr. Greg the following experiment: if we add to the cane juice a quantity of lime hardly higher than that which is necessary for the neutralization, the extraction with chloroform makes it possible to isolate the essential oil without other residue; but if, starting from this lower limit, the dose of lime is gradually raised, the chloroform ends by dissolving, besides this essence, some bad smelling products which are doubtless compound ammonia.

The importance of this experience cannot be over-emphasised, not only from the point of view of rum but also from the point of view of sugar production. At the Congress of Sugar Chemists held at Douai in March 1898, Mr. Hignette presented a memoir on the cold defecation of raw juices by a small quantity of lime; carbonation in a special turbine allows impurities to be quickly eliminated by the sole effect of centrifugal force. A few months later, Mr. Hignette himself experimented with his process on cane juice: the addition of lime to the dose of 2 thousandths of the cane juice used, followed by centrifugal carbonation gave it a very strong clear juice, purer than by ordinary processes (1).

(1) Bulletin de l’Association des Chimistes (Avril et Mai 1898)

The chemical study of cane juice therefore still presents many points to be elucidated and yet it precedes and dominates all the improvements that can be made to the sugar and rum industries. We have seen here only two sides of the question, the selection of yeast and the effects of lime. Other more recent works deserve as much development and I may come back to them later.

I point out, for example, the research of Mr. Edmond C. Shorey (2) who, studying the nitrogenous matters of the cane, recognized the existence of lecithins and glycocolle. Lecithins are very complex combinations which include compound ammonia, fatty acids, phosphoric acid and glycerin. They are found in abundance in egg yolk. They decompose under the action of heat into fats and alkaloids, slowly into a neutral liquid, very quickly into an alkaline medium. These bodies will therefore be found in cooked masses, where they hinder crystallization, and by extension also in molasses.

(2) Idein (Nov. 98)

Glycocoll is a simpler nitrogenous material, from the group of compound ammonia. This is the first time it has been reported in the plant kingdom. It is decomposed on boiling only by a large excess of lime and must be found almost entirely in the cooked masses. Its influence on crystallization is perhaps slight, but the aluminoid materials at the expense of which it is formed and which accompany it in the cane must play an important part in the formation of molasses.

I also mention the researches of M. Raciborski (Sugar cane 1898 p. 474) who was concerned with studying the causes of the rapid oxidation of cane juice exposed to the air. It is a fairly common occurrence that fresh sections of many plants turn brown quickly. This browning is due to oxidation by means of substances called oxidases which would perhaps play in plant physiology the role of blood hemoglobin in animal respiration. Mr. Raciborski has recognized two oxidases in the cane: one ceases to act from 60° while the other, which he calls leptomine, continues to act up to 95°.

The works that I have just analyzed show all the importance of the research which continues today on the Chemistry of Rum, and, if I could think that these studies are of interest to the readers of the Bulletin, I will not neglect to keep them informed of my personal research and of everything that will be published on the subject.


[Jamaica was the only region in the supplement with this many paragraphs regarding its rum production. I can’t figure out what all the reference numbers point to. When you go back to 111, there is nothing related to rum or Quentin Hogg.]

Fermentations. 403. To prepare it, molasses is distilled after having been fermented. The fermentations have been studied by Mr. Greg H. P. 41, 42, 43, who examined and selected the yeasts. General results which he furnished show that certain yeasts can completely ferment molasses in 1 day and a half, 2 days, 4 days, etc. 15 days. Rapid fermentations give more alcohol, but the rum they produce is aromaless. Slow fermentations give more aroma and less alcohol. Yeast No. 4 provides the most alcohol, and yeast No. 18 the best aroma. Mr. Hart 69, 70, criticized the conclusions that Mr. Greg drew from his experiments, although he recognized the value of the latter’s work. He disputes that Jamaica produces rum using selected yeasts.

[Hart was a scientist and journal editor from Trinidad who may have actually enticed Greg to move to Trinidad.]

German rum. 405. She makes, he says, mostly German Rum. This German Rum is obtained by a very special process which consists in preparing the fermentable material by letting it stay in a tank containing bagasse trash cistern. Distillation then gives a product very charged in aroma, which the German merchants buy on the spot, and which obtains the highest prices. Mr. Quintin Hogg can tell us that 111 these rums are sent to Bremen where they are mixed with neutral spirits in the proportion of 1 to 7. Products are thus made which are then sold as rums. Moreover, these same alcohols, doctored by chemicals are also used to make Brandy and rums where not even 1/7 of the Jamaican product enters.

[Quintin Hogg has an interesting reputation and among very many activities in the world of sugar, he introduced the Coffey still to the West Indies.]

Tasting. 403. Although able to appreciate by tasting the different eaux-de-vie and rums, I could not find any difference between the most vulgar rums of Kingston, and those which were offered to me as being superior. Their taste, which has nothing unpleasant, invariably reminded me of the article decorated in France with the name of Rum.

High degree rum. 406. This may be because the various sweets produce rum at a very high degree. This rum is sold wholesale to a small number of large merchants in Kingston who put it into everyday consumption after having subjected it to blending and wetting which undoubtedly unifies the taste and lowers the degree to around 60°.

2 thoughts on “G. Saussine—The Chemistry of Rhum, 1899

Leave a Reply

This site uses Akismet to reduce spam. Learn how your comment data is processed.

Discover more from Boston Apothecary

Subscribe now to keep reading and get access to the full archive.

Continue reading

search previous next tag category expand menu location phone mail time cart zoom edit close