Follow along: IG @birectifier
[I am re-editing this document after first posting it nearly 6 years ago.]
This article was published in the 1871, The Sugar Cane (page 191). The author hints are having written another paper, but it was likely published in a Jamaican newspaper.
I will easily say that it is one of the most significant papers on the topic of Jamaica rum and now that consensus has been built around many ideas, becomes a major pivot point for technique. It is a ten page process origin story that draws from Neilson’s European experience, the opinions of the German buyers with anecdotes galore.
Neilson tells of the mythic rum canes, early ripening canes having more aroma, and Bourbon cane producing his favorite rum. The German buyers paid the most for the first five batches of the season and we find out why. There are admissions of “lucious fruits”, the likely instance of a pombe fermentation, small estates doing it better. Read it for yourself.
Patrick Neilson may have worked at numerous estates, but who can claim him in 1871?
ON THE MANUFACTURE OF RUM IN JAMAICA.
By Patrick Neilson, Trelawney.
It is well known that Jamaica has always occupied the first place for her rum, owing probably to many causes combined, such as the high cultivation carried on, the peculiar properties of the soil, and in many cases from the superior manipulation of the material in the distillery. Many estates in this island make a fine rum, almost naturally, without any effort on the part of the bookkeeper to produce flavour, and in such instances of course it is better to let well alone; but it will be my object to show that on almost any estate it is possible to improve the rum, and I have had very conclusive evidence on the point.
To begin with, I admit that the soil has a great deal to do with the characteristic flavour of the rum of each estate, at the same time many soils may possess the same ingredients and properties, but they do not appear to be equally soluble, and refuse to part with their constituents, or yield the desired flavour by the usual processes, hence they require to be forced or coaxed, and stronger means employed to extract their virtues. To attain this, my chief object throughout this paper will be to show the great improvement that is obtained by adopting a system of slow and protracted fermentation in stillhouse operations, coupled also with a cool temperature; quite the opposite to the system usually carried on, namely, the hurrying in of all material, keeping the vats boiling like a pot, fermentation over in three days, closed doors and a heated apartment, consigning all the refuse and bottoms to a dirty cistern, where all, or at least a considerable portion is in a manner lost.
[The heated apartment idea may refer to the idea that some distillers started fire places in the fermenting house at the beginning of the season. Who knows how hot the temperature was elevated and if it had any baring on the microbiology of yeast selection.]
In order to show the great difference that exists between quick and slow fermentation we have only to look at the method pursued in some parts of Germany for the production of what is known as Bavarian beer, which is the product of a very slow and protracted fermentation, combined with a low temperature, for a full description thereof I would refer to Liebig’a works—a few extracts here however will serve my purpose so far, he says in regard to the manufacture of this beer, that “the wort after having been treated with hops in the usual manner is thrown into wide flat vessels, in which a large surface of the liquid is exposed to the air, the fermentation is then allowed to proceed while the temperature of the chamber in which the vessels are placed is never allowed to rise above 50°. The fermentation lasts from three to six weeks, and the carbonic acid evolved during its continuance is not in large bubbles which burst upon the surface of the liquid but in small bubbles like those which escape from an acidulous mineral water, a great deal of yeast is deposited on the bottom of the vessel in the form of a viscous sediment; this precipitated yeast does not excite ordinary fermentation again, and when the yeast is used again several times in succession the temperature may be increased much higher.” Liebig attributes the changes going on to a sort of putrefaction, and says, “that this remarkable process of fermentation with the precipitation of a mucous-like ferment consists of a simultaneous putrefaction and decay in the same liquid, the sugar is in the state of putrefaction and the gluten is that of decay.”
The beer so obtained is entirely different and far superior to many other sorts; and if this superiority is got simply from a slower fermentation then I hold that like results may be obtained in rum making.
[Jamaica had already valued fermentations that were not too quick relative to other traditions, but this degree of protracted ferment is on another level.]
I crave special attention to the foregoing, for it will be found to bear very much on the subject in hand, for in the precipitates, or bottoms of the rum fermenting vats lie one of the chief secrets in rum manufacture. The treasures that lay there looked upon as refuse, used in a manner to be thrown away.
[In the common clean process vat bottoms were always removed and moved to the dirty cistern or mud vat.]
What in fact first drew my attention to seeking out a flavour in rum was the running I first got from what is known here as the dirty cistern, a receptacle for all the refuse, bottoms, &c. of the other vats; I was astonished to find at the can pit mouth, as the rum came over, an exquisite flavour. I considered and said, why not make all the other vats, dirty cisterns, or at least to some extent, and I proceeded accordingly, with very happy results. I may state that the said dirty cistern besides being a receptacle for all refuse, is allowed to go on fermenting away, sometimes dead, sometimes alive, for three to four weeks; in fact, a sort of putrefaction goes on, and an acidity is produced which frees the aroma, hence the fine flavour when full.
[The dirty cistern or mud vat is a precursor to the muck hole and is used in a different way. It simply gets filled up then carefully distilled without any specific indicators of being ready besides being physically full.]
Following up my ideas, I then proceeded to turn half of the vats, gradually increasing up to two-thirds, into slow fermentation, and commenced by adding material in small quantities to each vat. The vats were about six feet deep, holding about 2000 gallons. I only put in liquor to cover about two feet at the commencement and added from day to day as I saw them dying off, keeping them always gently excited till they were filled up. I kept them moving this way for two weeks, sometimes less, sometimes even more. I set up at the rate of from 7½ to 10% of sweets, including skimmings, molasses, and dunder, which I kept at about 30° [Likely Long’s saccharometer so divide by 7.5 to get 4° Baume or SG 1.030]. I always used fresh water, unless of course I got copper water from boiling house. I sometimes allowed the molasses and skimmings to ferment together before introducing them into the vat. Molasses ought to be some weeks old before using at all. As these vats began to work I soon found a rich fruity odour pervading the house, and I then knew I had the right thing, which was confirmed at the can pit mouth. On resetting up these vats I did not as customary, pitch the bottoms into the dirty cistern, but set up again, on them, continuing to do so for three or four successive occasions when the liquor began to get muddy and I had to throw them out. I found at about the third setting up, a round from which I obtained the finest flavour, after that the vat began to get too heavy and sluggish and besides acid. I then threw the whole away, dunder included, and drew from the other third, that is, I kept one-third of the vats working on the old system, for the sole purpose of obtaining a fresh and sweet dunder.
[Notice at the end that Neilson is still concerned with excessive acid. I’m not positive what scale the 30° refers to, possibly Arnaboldi, but it would be helpful to convert that to SG.]
By setting up on these bottoms I extracted all their virtues, not so by a dirty cistern, it is far too clogged up, and there is not half enough liquor in it for the solubility of the precipitates, there is a loss there also, as the dunder has to be thrown away at every running of a dirty cistern.
By thus working I changed the character of the fermentation very soon, only small bubbles working on the surface quite gently, something like an acidulous mineral spring.
In carrying out the above system great care and discrimination is required by the distilling bookkeeper in order to prevent undue acidity getting into the house, for I do not deny but that the above results are obtained not only by a slow decay or putrefaction going on, but also by an acid generated thereby, and which acid assists in drawing out and setting free the good qualities of the bottoms and materials. When I wrote my first paper on the subject it gave rise to some comment—why could I only make two-thirds of my rum crop good and not the whole? The reason was I had to keep the remaining third working on the common system to produce the sweet dunder. Besides that, I then quietly withdrew all my low wine stills. I was the first to do this, but it has now become a common practice on many estates. I would however suggest that these low wine stills be redistilled—it would take away the low winey flavour.
[Even though this appears to be a free for all, undo acidity might be undesirable lactic acid bacteria which produces acetic acid which can halt a ferment. Neilson notices that desirable aroma may be a byproduct of acid creation. What I think Neilson is doing is re-distilling his heavy ferment low-wines to clean them up a bit, and then putting them in front of his clean ferments so that his clean process distillates draw down all the aroma. I have referred to this in the past as “juicing up the retorts”. He does not seem to be aware of concepts like acid catalyzed esterification.]
In addition to what has been said, there are many minor details requisite for procuring flavour, none more so than getting plenty of Rum Cane liquor, that is the half rotten or rat eaten cane, unsuitable for sugar and generally sent to the still house, not at all for flavour, but from economical motives, however the bookkeeper often gets flavour thereby quite unwittingly. The rum canes have gone through a process themselves of a slow fermentation, lying in the field and yard, hence their efficacy. I have known some estates making rum one year worth £25 in England, and the next only getting £17. I could trace it to the want of rum cane; when rum cane is not to be had, sweet cane liquor is as good only subjected to proper treatment.
[This begs the question, what is the current state of rum canes? And what is proper treatment? Is that simply passing cane juice over rat eaten rum cane trash?]
Another indispensable article is skimmings; a distiller cannot get too much of these. I have seen some estates giving away the half of them to the working mules, they were very probably throwing away £5 per puncheon; very few mules now enjoy that beverage, the value being better understood. The skimmings are the fat or cream of the liquor, and it is for want of that commodity that our large distillers in the United Kingdom can only distil guid Scotch whiskey out of molasses, in place of rum. Skimmings are a powerful agent in fermentation, and vats set up with molasses alone will not yield nearly as much as when mixed with a proportion of skimmings. Skimmings have a very acid reaction.
[Other authors have described skimmings as treated with lime, and how they should not be full of acid, but here we see acid which may be part of the pivot from high pH common clean ferments to high acid ferments. This may also be part of the pivot from acid skimmings to cane vinegar.]
I believe that temperature has something to do with making a good spirit,—”heat yer maut slow and ye will get the sweeter liquor” is an old Scotch saying, perhaps applicable here; it may be owing to this that our October brewing of ales at home are always the best. I often thought that in the beginning of the crop here, that is after the first round of vats, that the best rum was made, owing to the temperature—January being very cool. I have also heard German rum buyers here say that they have, engaged whole crops of rum on the faith of the first five puncheons submitted to them, but towards the end of the crop they were miserably disappointed.
It has often been remarked that the rum crop is made in the boiling house, and there is some truth in the remark, for stillhouse work is much dependent on what is going on there. Many a bookkeeper is charged with incapacity, for making a small rum crop, when in reality the fault was in a paucity of material. Nothing changes its character so rapidly as cane juice, and in proportion as it boiled off slowly or quickly so will the quantity of the material be affected—the more slowly the juice is evaporated the more fructose or grape sugar will be formed, which is molasses or near akin. Canes also, scarcely ripe, will produce much more for the stillhouse, and, bye the way, flavour also; even cane tops in a dry season are a very good thing for rum—just as good as rum cane, sometimes better. There are many other small matters in the boiling house, all tending either for the benefit or detriment of the stillhouse, such as the amount of temper lime applied in the juice, boiling high, potting cold, &c., &c. In one estate I was on, I saw lime used and was ordered to use it myself to a very injurious extent; on other estates I have been on no lime was used at all, and very proper too, for at all times a perfect neutrality must be observed. An excess of lime used in whitewashing vats, &c., will of itself produce a virulent acid, something of the nature of lactic acid, and, besides, destroys fermentation.
[At the end here, we see mention of the whitewashing technique of dry rubbed lime used to clean vats and gutters. Very likely it can lead to too alkaline pH that promotes the growth of undesirable bacteria. It seems like a skill of the common clean process is understanding when you have crashed and starting completely over.]
What also affects cane juice, but in favour of the stillhouse, is the distance sometimes that the grinding apparatus is from the coppers, especially where there are windmills, the liquor has often to travel 200 yards in an open gutter. That the manner of boiling cane juice does affect after products is shown by the fact which I have heard stated on good authority, that where on an estate usually producing a good rum, they were compelled, owing to an accident in their machinery, to send their canes to be ground in a neighbouring estate, the rum turned out perfectly different.
What sort of cane is most favourable for good rum it would be difficult to say, but I certainly prefer the Bourbon.
The proportion of rum made to sugar fluctuates very much in Jamaica; however, three-fourths, that is, 75 puncheons of rum to 100 hhds. of sugar of 1 ton, is considered a fair proportion, many estates do not make above half. It is very evident, however, that if the yielding in the boiling house is as low as 1200 gallons to the hhd., there cannot be much going the stillhouse way, all is being converted into sugar; on the other hand, if the yield is about 2000 gallons, longer time being required for evaporation, much more molasses and grape sugar is generated. Only observe how Fryer’s Concretor works; from the very swift process no molasses is formed. There are many ways of making a large crop of rum in proportion, one I need scarcely mention, which is known to many distillers, but the spirit is very thin and poor.
Some bookkeepers manage to make a large crop by being on very good terms with the boilermen. I was once on an estate with a very clever bookkeeper who was making puncheon for hogshead, mirabile dictu; the yield was 3000. Some estates making fine rum send down liberally from the boiling house in order to increase the quantity of rum, but this is easily overdone, for although cane juice alone, if properly treated, is of great value for flavour, sugar has quite the contrary effect. I knew one estate whose rum commanded the highest price in the English market, where they sent down the greater part of the sugar one year; the brand of the rum also went down to an alarming extent.
I have seen it stated on high authority that 1 cwt of sugar will make 10 gallons of spirits, but I think that this is a mistake, for in that case all our estates would be converting their sugar into rum as being more profitable, for supposing that 1 cwt. of sugar would yield 10 gallons of spirits, then 1 hhd. of 20 cwts. would give 200 gallons or 2 puncheons rum, valued say on the spot at £24, whereas the 1 hhd. sugar would only be worth £15, thus there would be a profit of £9, barring a very little for the manufacture of rum.
I have also seen it stated, that 1 cwt. of sugar will yield 10 gallons of spirit, and 1 cwt. of molasses, 7 gallons; now, I always found it just the reverse; more, I would say that either weight for weight, or bulk for bulk. Molasses will yield three times the spirit that sugar does.
Leaving the fermenting house, we come to the last process in rum making, namely, the distillation of the fermented wash or liquor, this is so simple that no description is necessary. The stills in use here are mostly those having one or two retorts, which produce rum at one distillation, there are a few estates yet using the old single still, which involves three distinct distillations and great extra expense for fuel and attendance. It would be difficult to say which still produces the best rum. One would imagine that with the retorts great opposition is offered to the essential oils, or flavor coming over, and yet some of our best rums are obtained by retorts.
I have seen some bookkeepers using the spent lees of the retorts again, saying they get better returns; there never was a greater fallacy, and of all things I would eschew this, the low winey poison is just what I would seek to avoid. It is amusing to hear the different opinions about loading the retorts, the fact being that as long as the induction pipe is covered it is all the same, the alcohol will come over, it may be either in rum or low wines. To save trouble it is best to keep strong spirit in the retorts, and when kept up a better spirit is got. Consign the weak rum to the low wine cistern.
[I’m pretty sure implies mixing the spent lees from the retorts back to the ferments, but the acids are known to be antagonistic to the yeast as well as aroma beneficial bacteria in the dunder. “Better returns” seems to typically imply returns on sugar, a clue being “low winey poison”, and not distillation efficiency or size of hearts fraction.]
There are various opinions as to whether rapid or slow firing is best, I always prefer slowish firing. It must be in borne in mind that although the boiling point of alcohol is 173°, whilst that of water is 212°, that an intermediate temperature is requisite to bring over the spirit, depending on the strength of the wash or charge, in the still. The following table shows approximately the boiling points of wash or fermented liquor, containing various proportions of alcohol, and also the proportions of alcohol to water in the vapour arising from such liquids and distilling over.
As the charge in our stills in this country never exceeds 13%, it will be seen that the closer the firing is kept to the boiling point of water, the better. Some think, however, that with strong firing the essential oils come over better, especially in the case of retorts.
[More low volatility, high molecular weight aroma can come over if you apply more energy to the boiler because you challenge the natural reflux of the pot still and distill at a relatively lower ABV. This also relates the “k’ concept” where some high molecular weight aroma distills in the heads at a low ABV but switches to the tales above a certain ABV like 30%. As a metaphor, it is like on a motorcycle where you are steering at low speed and counter steering at higher speed. When the reflux is challenged by adding more energy, they may be at a threshold where whatever steering they are used to is inverted. Many people have tried to harness this effect, but I do not believe there is any aroma beneficial ways to do it because of aroma-negative side effects.]
I have seen Coffey’s apparatus working at home, it is a beautiful and highly scientific contrivance, based upon the known principle of alcohol separating at a certain temperature.
The fact has already been noticed, that when spirit is mixed with water it does not boil, or rapidly vaporize until the mixture is heated to a mean temperature, depending upon the proportion of water and spirit present, and then both water and spirit vaporize together, and also the essential oils which may be present, it must now be noticed, that if instead of immediately condensing such mixed vapour, it is allowed to diffuse itself in a chamber maintained at a temperature a few degrees above the boiling point of alcohol 173° Faht., nearly the whole of the water and oils will resume the liquid form, carrying with them but a small portion of the spirit, and the spirit vapour may then be drawn into another chamber at a lower temperature where it will be condensed, and spirit will thus be obtained in a state of great purity. Upon this principle the process of rectification is conducted in Coffey’s apparatus.
For a full description I beg to refer to the patent itself, as it would take up far too much space here: I may briefly say that the apparatus is a chamber built of wood, about 8 feet high and 4 feet square, this is divided into compartments, or shelves made of copper plates perforated with holes. The wash is introduced at the top, and steam from a common steam boiler is introduced at the bottom and plays upon the wash in its descent downwards, and by some very ingenious internal arrangement spirit as strong as 65 Sikes [94.28% ABV], comes off one way, and the spent wash goes another, there are no such things as high and low wines. I do not think that such an apparatus would do for very small estates, but might be used advantageously on large estates. It is intended, however, for a pure spirit, and this is not desirable for rum, only this might be avoided by raising the temperature of the chamber where the steam is passing through, so that the essential oils would come over also, as it is, the essential oil is collected separately and used for flavouring brandies and other spirits, why not also the rum itself? Certainly the process would improve those rums which have a bad flavor, for it would take it away. Where estates have already steam boilers it would be the more applicable. I suspect the apparatus would be cheaper than the stills in use, there would also be great saving in fuel, and as the spirit can be, made at 65 [94.28% ABV], in place of 35 Sikes [77.14% ABV], there would be the saving of half the puncheons, which calculating at 30s. each would be £150 in 100 puncheons, in freight also at £2 per puncheon there would be £200, in all £350.
I will close my paper with a few general remarks on the subject. It has been observed that few very large estates make fine rum— always the smaller ones. I account for this by the bookkeeper being obliged to keep pace with the large crop, and in consequence, hurry on all his materials. Molasses, for instance, should not be used till they are a week or two old, they oxydize as it were, and certainly change their character. Perhaps it is from other causes that our “wee still whiskey” at home is famed.
Another important matter that will affect rum is the quality of the water used. I believe that the name of some of our more celebrated brewers and distillers in England depends upon this.