Selected Writings of Fermentation Chemist S.F. Ashby

More alcohol was added to c, d, and e, after three weeks, and the acid rose to 6.2, 5.8, and 5.3, respectively in another week, but showed no further increase. The greatest amount 01 acid produced was therefore equal to about 7.5 per cent, of pure acetic acid, the largest quantity which the bacterium could endure. The organism could not grow and work in 24.7 per cent of proof spirit, and showed only feeble activity in 16.5 per cent, but in 12.3 per cent, it worked strongly. The evidence shows therefore the amount of alcohol which can undergo vigorous acidification is between 12 and 16 per cent proof spirit, which agrees with the result of the first observation.

The theoretical maximum amount of acetic acid which could be formed from the alcohol in cultures c, d, and e, is 7.3, 5.9, and 4.9 per cent. The actual amounts formed in 20 days were 5.7, 5.2, and 4.6 so that
in c 78 per cent of the possible was formed,
”  d 89
”  e 94

The lower the amount of alcohol in a liquor, the more completely therefore is it oxidised to acetic acid. For practical purposes the highest acidity was reached in a fortnight at about 4 per cent. Bacterium II. proved to be unable to grow and produce acid in a dead wash containing 12 per cent proof spirit, but gave over three per cent acid in a liquor with 8 per cent proof spirit. This bacterium also makes greater claims upon the nitrogenous foodstuff in the liquor than bacterium I. Bacterium I. is therefore the characteristic acetic acid producer in all liquors containing 10 per cent and more of proof spirit, such as ordinary dead washes, while bacterium II. works best in liquors like fermented skimmings and fermented rum cane juice.

The following table shows the amounts of Total and Volatile acid (mostly acetic acid) and the relative amounts of volatile acid to total acid in some distillery liquors. Of special interest are the quantities of volatile acid in such materials as acid skimmings, and flavour, because in these liquors an attempt is made to produce as much volatile acid as possible. The volatile acid shows an average percentage of the total acid of from 22 to 27, or only about one quarter of the acid present is volatile. As the fresh skimmings which comes down from the boiling house are practically neutral the great part of the acid produced in the cisterns is the work of bacteria. Although the skimmings readily undergo fermentation, this is not entirely due to yeast, as the liquor is heavily contaminated by bacteria which produce fixed acids such as lactic from sugar. A number of such bacteria have been separated from the skimmings. They include the well known rice grain bacterium, which can nearly always be found in skimmings. It forms large rounded gelatinous masses when strongly developed consisting of enormous numbers of hand shaped colonies, the rod shaped bacteria being embedded at the ends of finger like processes of the jelly. This bacterium produces lactic acid and forms its jelly at the expense of the sugar present. Another rod shaped organism often develops in fresh cane juice contaminated by dirt from the mill or by soil, at a great rate, and converts the liquor in one day into a thick viscous mass in which yeast can only work very sluggishly. Gas and lactic acid are produced, the viscous substance being formed at the expense of the sugar. The presence of such objectionable organisms account for the poor yield of alcohol in skimmings, and the small amounts of volatile acid. Acetic acid bacteria are wholly dependant upon oxygen for their work of converting alcohol to acetic acid, and require therefore that the liquor in which they are working should expose as great a surface as possible to the air. This is only being imperfectly attained in distilleries even in the trash cisterns. It is proposed therefore to Experiment on a practical scale with a view to the more rapid and more abundant production of acetic acid from alcoholic liquors.

 

PART V.
REPORTON THE STUDY OF FERMENTATIONS IN THE MANUFACTURE OF JAMAICA RUMS.
By S. F. ASHBY, B.Sc, Fermentation Chemist.

1. Useful Information’ Regarding Estate Distillery Materials.

Skimmings or Scummings—A mixture of liquor and solid matters skimmed from the surface of juice in clarifiers and coppers (if used) together with wash water from coppers, etc. The solid matter a mixture of pulverised cane fibre (trash), phosphate of lime, pectic and waxy matters, and coagulated albumen. According to the amount of solid matter and of dilution the gravity may vary when quite fresh from that of the juice (15-20 Brix) to under 10 Brix. The reaction to litmus is either neutral, faintly acid or faint alkaline.
Dunder—The liquor left in the still after distillation is completed. A yeast extract. The gravity varies according to materials fermented from under 10 Brix to over 25 Brix, and the same applies to the acidity which varies from about 1 per cent, to over 3 per cent. It is never free from sugar which varies from 0.2 per cent, to over 1 per cent. Sugars other than hexoses (pentoses) and allied bodies may be present which reduce Fehling’s solution but are not fermentable by yeast. On an average about 1 percent, of glycerine has been found in Dunder. It is never free from volatile acid.
Its high density is due to cane and yeast gum and caramel (especially if still is direct fired.)
Molasses—The sweet viscous syrup separated from the crystalized sugar by the centrifugals. It is markedly acid (about 0.5 per cent.) has a specific gravity of about 1.45, contains about 40 to (50 per cent, cane sugar, and 10 to over 20 per cent, glucose. One gallon (imperial) contains 8-10 pounds of fermentable sugar.
Acid—Skimmings, normal cane juice, or rum cane juice, allowed to sour. The production of acetic acid is the object sought. The volatile acidity rarely exceeds 40 per cent, of the total and is usually under one third the total.
The souring is carried out either with trash cisterns or without the addition of trash. The liquor ferments (yeast) and sours simultaneously.
Lees—The liquor left in the retorts after distillation is completed. It contains a high proportion of volatile acid.
Wash—The liquor (prepared from the mixed materials) which is actually fermented and distilled for rum. The mixing of the materials is called “setting up.” When fermenting it is “live” wash, and when fermentation has ceased it is “dead” wash.
Flavour and “Muck Hole”—(See description in first Sugar Experiment Report.)
Rum—The early portion of the alcoholic distillate; (the preliminary runnings if cloudy are rejected) its strength varies from 36 to over 40 proof as determined by the “bead.” It is water clear (white Rum). Before leaving the estate “Rum store” it is coloured by caramel boiled by the distiller. Each estate has its own standard of colour.
High Wines—The running from the still which follows the rum; collected to a strength of about 20 over proof.
Low Wines—The subsequent runnings collected till all alcohol has distilled over. The strength varies from 40 to 60 under proof.
Retorts—Copper vessels inserted between the still and the coil. The vapours from the still must pass through them. Most estates have one retort which contains the high wines of a preceding distillation. Some estates have both ”high wines” and “low wines” retorts, the latter next to the still. The retorts have a capacity of about 1-10 that of the still.
Low Wines Rum—Some estates with one retort (high wines) add the low wines to the wash in the still; other estates, however, distill the low wines independently (they run about one low wines still to 5 or 6 ordinary wash stills) and obtain “low wines rum” a product of inferior quality and price.

Types Of Rum.

The two main kinds of Rum are “Common Clean” and “Flavoured or German.” The individual estates confine themselves to the manufacture of one of these kinds. Nearly all the “Flavoured” Rum is made in the parish of Trelawny.

Common Clean Rum—may be divided into two kinds depending on the materials used.
1. From washes set up with a mixture of skimmings, dunder, molasses and water. The materials are not allowed to sour. Several estates with up-to-date boiling house plant (vaccuum pans, etc.) and a consequent large out put of skimmings and molasses employ this method. The materials must be used rapidly, and fermentation rendered of as short duration as possible. The wash is set up with 1/3 skimmings, 1/3 dunder, and molasses and water to give an initial gravity of about 10 Brix. The wash attenuates in about 4 days to 3 or 4 Brix. The initial sugar content is about 11-13 per cent, and the attenuation from 11-13 degrees. The rum is light in body and of low ether content, and is mainly consumed locally.
One or two estates which do not make sugar boil their juice and ferment it with dunder. (Appleton).
2. From washes set up from the same materials and also with “acid” prepared either from skimmings, rum cane juice or normal cane juice. The composition of the wash varies:—

The gravity of the setting depends largely on that of the dunder which varies from 10 to 20 Brix. As a rule the setting is not lower than 18 Brix. and may be as high as 24 Brix. The initial sugar content varies from 10 to 14 per cent, and the attenuation corresponds to that. The fermentation period depends on both the acidity of the dunder and on the quantity and acidity (especially the volatile) of the “acid.” The wash ferments from 5 to 9 days and is often allowed to lie for a couple of days when dead.

The only acid produced is evidently “acetic” and some of these rums may contain over 1,000 ethers (Swanswick, Long Pond) where much “acid” is used in the wash.

The yield of proof spirit is from 0.85 to 1.0 per cent, on the sugar fermented and on the attenuation 0.8 to 0.9 per degree. From 5 to 10 per cent, is lost in distillation.

The yield of rum 40 o.p. varies from 60 to 90 gallons per 1,000 gallons wash in still.

The fermenting cisterns (sunk in floor of distillery built of wood and backed by puddled clay) and vats are usually of 1,200 gallons capacity and the still will receive the contents of one cistern. Two stills are usually run per day (daylight). The stills are heated by steam coil or by direct fire. The rums made with ‘common clean’ materials vary in ether content from under 100 parts to over 1,000. Acetic ether is practically the only one present, and its amount depends entirely on the quantity of acid used in the washes and on the length of time the wash ferments and lies when “dead.”

Flavoured or German Rum.—These rums are made on estates having old fashioned boiling house plant where the manufacture of sugar is of secondary importance. The usual common clean materials are employed and in addition “flavoured.”

“Acid” is prepared from cane juice or skimmings in the usual way in a succession of trash cisterns. A “muck hole” outside the distillery is the receptacle for the thick matter deposited from the dunder, and the wash (dead wash bottom) to which is added cane trash and lees. The matter consists to a large extent of dead yeast and is therefore highly nitrogenous. It undergoes slow fermentation and putrefaction and its acidity is kept low by the addition of marl. When ripe it contains large amounts of butyric and higher fatty acids, both free and combined with lime. It is added to a series of acid cisterns outside the distillery where the butyric and other acids are set free. This complex acid material is the “flavour.” The flavour enters the wash after fermentation has begun owing to the presence of acids in it which are injurious to yeast, the fermentation is prolonged and the sugar is never very completely fermented out. Fermentation lasts 9 to 10 days and the dead wash lies for several days longer. An example of the kind of wash follows:—

This means a yield of 48 galls, rum per 1,000 galls, wash whereas the attenuation would indicate a yield of about 78 gallons. Only a portion of the high strength distillate is therefore collected as rum of first quality.

These rums show an ether content as a rule from 1,000 to 2,000. While over 95 per cent, of the total ethers is “acetic” there is always present several per cent, of butyric ether and still smaller amounts of esters of higher fatty acids (capryllic, caproic and lauric). Most of these rums find their way to Germany for blending and particularly for “stretching” potato or molasses spirits.

MlCRO-ORGANISMS OF THE DlSTILLERY.

Yeasts.—Practically three yeasts perform all the conversion of sugar into alcohol in the Jamaica Distillery.
1. Bottom fermenting oval budding yeast.
2. Top fermenting chained fission yeast.
3. Bottom fermenting unchained fission yeast.

Oral budding yeast.—A typical bottom fermenting yeast the cells of which do not form chains. It is oval in shape and often rather pointed at one end. The average dimensions are 7.5-9 m long by 6-7 m. broad. It does not form a film on dead wash but at most a yeast ring. It forms spores on the gypsum block (as a rule four in a cell) in 24 hours at air temperature. It readily inverts and ferments cane sugar. This yeast is present on the rind of the cane and is always found in freshly milled juice. Spontaneous fermentation of juice is therefore always brought about by this yeast. In fresh juice it multiplies quickly and sets up a rapid fermentation. It displaces all other native yeasts in a favourable liquor like juice. The optimum temperature for its multiplication lies above 30 C. but it appears to ferment best at that initial temperature. It will work practically all the sugars out of an undiluted juice if not interfered with by acid-producing bacteria. The fermented liquor has an agreeable odour. In the experimental work at the Sugar Station Distillery where either cane juice or cane juice, molasses, and dunder are usually worked with, this yeast alone sets up and carries through normal fermentation.

On estates where the first type of common clean rum is made (i.e. without “acid “) this yeast possesses the wash owing to its properties of quick multiplication and rapid and intense fermentation. Such washes heat up quickly and temperatures as high as 108 F. have been observed. These high temperatures mean injury to the yeast, imperfect attenuation, and marked loss of alcohol by evaporation. Like most bottom fermenting kinds this yeast is markedly susceptible to unfavourable conditions such as poor food supply, excessive temperature and especially high acidity. Volatile acidity injuries it very readily (see experiments in second S.E.S. Report.)

It is injuriously affected by the fixed acids of dunder and works best where the initial acidity of the wash does not exceed 0.3 per cent. In washes with an initial acidity of 1 per cent and more it gradually gives place to more acid-resistent yeasts. On estates using acid the wash contains both this and fission yeasts, the relative proportion depending on the amount of acid employed. In common clean washes with an acidity exceeding 1.5 per cent, and a volatile acidity of 0.5 per cent, the writer found it entirely displaced by fission yeasts even quite early in the season.

Top Fermentiny Fission Yeast.—A typical top fermenting chained yeast. On washes of high acidity which are not working very intensely this yeast throws up a characteristic light or dark golden yellow thick moist creamy or fatty head which may completely cover the surface of the liquor. The bubbles of gas escaping through the head are cloudy. The head consists mainly of short, rectangular cells in chains of four or more, often in clumps and showing a kind of false branching. When shaken up in a wash the yeast forms into loose flocks which rapidly deposit. There is considerable variation in the size and shape of the cells: the size varies from 6-12 m by 4.5 to 5.5 m. and the chain cells are usually small viz., 6-7 m. long by 4.5 m. broad.

Spores are freely formed in the wash during fermentation. There are four oval spores in a cell and their walls stain blue with iodine (in iodide). The spores are very frequently found in bridge shaped sporangia formed by the reunion after division of two cells or by the union of two neighbouring cells. This yeast has a high optimum for multiplication and fermentation between 34 to 37 C. It endures high acidity (over 3 per cent. total) and is greatly more resistent to volatile acid than the budding yeast. At ordinary temperatures 24 to 27 C. the fermentation is slow but the sugar is efficiently worked out. In pure cultures the attenuation and the yield are as good as from the oval yeast.

In all washes of high total acidity (over 1 per cent.) and especially of high volatile acidity this yeast is generally present and often carries out the entire fermentation. It is the typical yeast of the “Flavoured Rum” washes.

Bottom Fermenting Fission Yeast.—This yeast produces no head in washes, the escaping bubbles being glassy clear. The cells are found single and in pairs, and when the wash is stirred the cells distribute themselves in a fine clay-like suspension, which clears slowly. The cells are variable in shape and size averaging 6-14 m. long by 4-5.5 broad. Spores are formed with the top yeast.

This yeast has a somewhat lower optimum temperature than the top yeast, and like most bottom yeast yields markedly less substance and is more susceptible to external factors than the top yeast. It is often found in acid washes together with the top yeast. It increases more rapidly and ferments more strongly than the top yeast at ordinary temperatures. The attenuation and yield of alcohol in pure cultures are the same as for the top yeast; it appears to leave more unfermented sugar in the dead wash. In a sample of soured cane juice having a total acidity of 2.1 per cent., and a volatile acidity of 0.90 per cent, this yeast alone was found. In the wash set up with this “acid” having an acidity of 1.6 per cent, and over 0.50 volatile, the top yeast was the characteristic worker. It would seem that in highly acid washes the more resistant top yeast gradually increases with the advance of the season. Comparative experiments with these two fission yeasts in Laboratory washes at air temperature indicate that the bottom yeast ferments the wash in one or two days less time.

A bottom yeast with slightly top phenomena has also been isolated. It forms no chains but the cells agglutinate more than the typical bottom yeast. In its properties it comes between the two extremes.

The undermentioned yeasts isolated from distillery materials play no evident part in the actual fermentation of washes.

Fruit Ether Yeast.—Isolated from ”foaming” molasses. Forms a dry white friable wrinkled film on material containing sugar. A small oval budding yeast with cells very variable in size. Forms spores on the gypsum block in 18-24 hours at air temperature. The spores are “hat shaped.” The yeast is therefore an “anomalus” variety (Willia anomala). It inverts and ferments cane sugar and will ferment diluted molasses over 50 Brix.

Produces a very high amount of acetic ether, the distilled wash containing from 12,000 to 40,000 ethers. The fermentation is slow occupying two weeks or more to attenuate 12. In ordinary washes it is easily displaced by more active yeasts. (See second S.E.S. Report, and experimental data in Laboratory records).

Pastorianus Yeast.—Isolated from “foaming” molasses. A top fermenting yeast which forms spores abundantly on the gypsum block in 18 hours. Will ferment diluted molasses of 50 Brix. Inverts and ferments cane sugar but is easily surpassed by the oval budding yeast and fission yeast in appropriate washes. It yields a fermented product of good aroma and has been used successfully in the preparation of orange wine.

Torula from Molasses.—Does not form spores and cannot invert and ferment cane sugar. A small chained oval or spherical yeast which ferments the glucose in molasses of the highest gravity. The cause of “foaming” (see second S.E.S. Report)

Large celled Oval Yeast.—A spore-forming top fermenting yeast which works badly in estates’ washes.

Ludwig’s Yeast—Found occasionally in small amount in fermenting cane juice and also present in “acid” (Long Pond and Swanswick). Is probably present on the cane. Corresponds in size, division, and spore formation to Saccharomyces ludwigii.

Mycoderma Species.—Commonly present as grey and white wrinkled films on dunder, sour skimmings, and dead washes which are allowed to lie two or more days. Produce no fermentation, but oxidize alcohol to carbonic acid and water. No spores formed.

Culture Media.—The medium which had answered well for the cultivation of all yeasts is a cane juice peptone broth. Prepare as follows:— Fresh cane juice is tempered with milk of lime, heated to the boiling point and filtered. Care must be taken to avoid excess of lime or the liquor will darken excessively. This liquor may be transferred to a Carlsberg Can and boiled for half an hour on two successive days to sterilize it.
The medium has the composition:—
tempered cane juice 100
peptone 0.5
potassium phosphate 0. 05-0.1

The cane juice should first be diluted to 13-14 Brix. The broth is heated in the steamer and filtered and then rendered distinctly acid with Hydrochloric acid or Lactic acid. The acidity should be between 0.05 and 0.1 per cent. If not clear the medium is allowed to stand a day and filtered again. It is distributed into Frendenreich flasks (a few o.c. in each) and sterilized by heating ½ hour in the steamer (Koch’s) on three successive days.

To prepare a solid medium 1.5 per cent, agar is dissolved in the broth. The medium should have a more or less pale sherry tint and should not be reddish brown.

Media containing dunder are very dark and difficult to clear. The acid of dunder affects the solidifying power of agar. 10 to 15 per cent of gelatin may be used instead of agar but cultures must then be kept in the cool incubator at 20 to 22 C. The yeasts grow much better in the cane juice medium than in a purely artificial one. The oval budding yeast retains its vitality well in the cane juice broth for over a year. The fission yeasts die out more easily but living cells are present in fair numbers after twelve months.

Fermentation experiments are carried out in flasks containing 1 litre of wash and plugged with cotton wool. Washes with an acidity exceeding .7 per cent, are generally sterile after ½ hour steaming. The progress of fermentation is determined by daily weighing, the loss being taken as carbonic acid.

The following factors are determined in all experiments (Laboratory or distillery).

The wash should be quite dead and the yeast well settled. If yeast is suspended the spindle gives a reading 0.15 to 0.4 too high. After 48 hours the reading will be correct. The weight of sugar fermented is very closely double the loss of weight.

On estates the Arnaboldi or Jamaica Saccharometer is still frequently employed. It is corrected for 80 F. and gives a reading roughly half as high again as the Brix spindle.

Sending Yeasts To Estates.

At the start of crop the distiller gets a spontaneous fermentation in whatever liquor he can get. This is either skimmings, cane juice of low gravity and purity (rum cane juice) or, if he is fortunate, fresh juice from the first mill. Dunder left from the preceding season is often used to mix with the juice. This dunder often contains matters which inhibit or interfere with the growth of the yeast. It is improved by vigorous boiling with or without the addition of lime. As soon as he has molasses and fresh dunder he can set up a normal wash. If he has to start on skimmings they often contain very little or very feeble yeast, and easily get spoilt by bacteria which render them ropy or viscous. Much difficulty is therefore often experienced in getting a good start to fermentation. In any case the yeast which develops is the oval budding kind. If “acid ” is made and used from the outset in quantity the oval yeast frequently works badly and gives place gradually to the more suitable fission yeasts. In the meantime there may be loss by bad attenuation and accumulation of materials.

In sending yeasts to estates at the start of each crop the object of the Laboratory had been to get in suitable yeasts from the outset and curtail the period of uncertainty. For estates not making “acid” the oval budding yeast, and for estates making acid—one or both of the fission yeasts.

Yeasts were sent to a few estates in December 1907, and January 1908, to still more in December 1908 and January 1909, and to over twenty estates in December 1909 and January and February 1910.

The following estates got yeast this crop:—

Yeasts are required from the middle of December to the middle of February. The estates taking them later had really started weeks earlier. The number of Common Clean Estates willing to test the yeasts could doubtless be doubled for the crop 1910-1911. They would need to be circularized in November.

Preparation Of Yeasts For Estates.

The yeast is first grown in the Frendenreich flasks in the cane juice broth. Two or three transfers should be made when fermentation has almost ceased; ½ c.c. suffices after shaking up the liquor. This is done with sterile pipettes in the glass chamber after washing down the latter with 2 per 1,000 mercuric chloride solution,; 5 c.c. are then added to 60 c.c. sterile wash in small Pasteur flasks. After fermenting in these flasks for 3 to 4 days they are shaken up and the whole liquor poured into flasks plugged with cotton wool containing 1 litre sterile wash. When fermentation has nearly ceased these flasks are shaken and the liquor poured into large flasks containing 10 to 12 litres sterile wash. The wash is allowed to die completely and the yeast permitted to settle out (24 hours after wash is dead). The covering liquor is then poured carefully away and only sufficient left to give a thick muddy suspension when the yeast is is shaken up with it. The mixture is poured on to moistened filter paper in Buchner funnels and the moisture drawn out as effectively as possible by means of the Geryk air pump. The yeast and the filter paper are lifted out, wrapped in dry filter paper with a covering of glazed paper, packed in a small tin with cotton wool and mailed by Letter Post to the Post Office nearest the estate without delay. The estate must be advised to get the yeast working on the day of arrival.

The washes in the Pasteur, litre, and large flasks should, for preference, be set up from a mixture of molasses, dunder and water, using 1/3 to 1/2 dunder (according to its acidity and gravity). The gravity of the wash should be such that it will attenuate 12 if allowed completely to die. The Pasteur and litre should have added to them 0.2 per cent, asparagin, and the wash in the large flasks 0.1 to 0.2 per cent, ammonium citrate or ammonium sulphate. In the absence of molasses muscovado sugar or concentrated cane juice may be used. If neither dunder or molasses are available the wash may be set up with muscovado sugar and citric acid (1 per cent, of a gravity of 12 Brix). To this should be added .2 per cent, asparagin for the Pasteur and litre washes, and .2 per cent, ammonium citrate for the large flasks, .05 per cent, potassium phosphate should also be added.

Before adding the yeast the wash should be warmed to 30 C. The litre and large flasks should be packed round with saw dust or better fibre packing to keep up the temperature and make the fermentation more uniform. The flasks containing the litre washes should be weighed daily to judge if fermentation is vigorous and normal. If a yeast is to go to several estates within a short period a little may be kept back in the large flasks after decanting off the dead liquor and this will serve to start another large flask (or more than one). Before sending away, a little of the yeast should be examined under the microscope to observe if it is true to its type and free from living bacteria.

Directions For Working The Yeasts On The Estates.

Set up ten gallons of fresh wash in a clean keg; the wash to consist of dunder 1/3 molasses and water, and to be of such a gravity as to give an attenuation of 12-13 Brix (18-19 Arnaboldi) if the wash were allowed to die completely. The temperature should be 86-88 F. Stir in the yeast, cover the keg and allow to stand in a warm place. When this wash has lost 9-10 Brix (14-15 Arnaboldi) by attenuation, stir up properly and pour the entire liquor into 50 gallons fresh wash. When this has attenuated to a like extent stir up and pour the whole into 500 gallons freshly set wash. When this is working well (after 24-36 hours) make up to 1,000 – 1,200 gallons. A freshly set 1,000 gallon wash can be started again from that by adding to it 50- 75 gallons when the attenuation has fallen 9-10 Brix and after thoroughly stirring up. The yeast can be got through the distillery more rapidly by keeping back 10 gallons of the fermenting 50 gallon wash and using it to start a fresh 50 or 100 gallon wash in the same puncheon (with the head knocked out) which may be poured into 1,000 gallons when it has attenuated 9-10 Brix. Skimmings should not be used in setting up wash except in the last 500 gallons.

When circularising the estates they should be asked if they propose to use “acid” in the coming crop. Content, Kent, Cinnamon Hill. Running Gut, Ironshore, Gale’s Valley, and Swanswick have already employed the top fission yeast with success. Catherine Hall, Albion and Parnassus would be best suited with the oval budding yeast. Sevens, Spring, Appleton, Denbigh, Bog and Llandovery, Green Park and probably the Belleisle Estate Co. might get both oval and bottom fission yeasts. Orange Valley should get both fission yeasts. If two yeasts are sent together the distiller should be advised to grow- them separately in 10 and 50 gallons and then pour the two 50 gallon washes together into 1,000 gallons of ordinary estate wash. The yeast better adapted to the conditions would then get the upper hand.

Bacteria.

Acetic Bacteria.—Forming a film on liquors containing alcohol oxidizing it to acetic acid. They can be isolated from dead washes, “acid,” etc., by means of cane juice peptone agar to which 2 per cent, of alcohol has been added. The commonest forms are B. kutzingianum (or an allied species), B. xylinum and B. xylinoides. The first named does not give the blue stain with iodine. It forms a blue to white delicate very friable ascending film and clouds the liquor strongly. (For experiments with this species see second S.E.S. Report and Laboratory records). B. xylinum develops the characteristic tough thick white skin on any nonfermenting liquor containing cane sugar and not less than 10 per cent, proof spirit.

B. xylinoides forms a very similar skin on “acid” and liquor containing over 10 per cent, proof spirit. One or more of these species are always present in fermenting cane juice or estate washes and cause a rise of acidity by forming gluconic acid from sugar. When the wash is dying and especially when it is dead they produce acetic acid.

Saccharobacillus pastorianus has also been found in fermenting washes and especially in soured skimmings and cane juice. It is present as long narrow rods often covered with small particles of matter deposited on them from the liquor. The liquor is strongly clouded and the bacteria cause the appearance of silky waves. This organism grows freely in the presence of alcohol and therefore increases with the yeast during fermentation. In cane juice broth it gave rise to 0.8 per cent, total acidity of which 30-35 percent, was volatile (acetic). The fixed acid is lactic acid.

“Acid. “—This is prepared either from skimmings or cane juice by allowing them to ferment and sour in special cisterns. As a rule trash is added to the liquor which on some estates is pumped into a succession of cisterns in each of which an increase of acidity occurs. The acidity of the ripe acid rarely exceeds 2.5 per cent, and of this as a rule less than 1/3 is volatile. The highest volatile acidity hitherto observed was 0.9 per cent, out of a total acidity of 2.1 or about 43 per cent. The liquor is fermented by yeast (oval or bottom fission) and the acid increases rapidly at the same time. This increase frequently stops attenuation when several per cent, of sugar is still present. Hence “acid ” shows a most variable gravity according to the relative activities of the yeast and bacteria. The amount of acid formed is the same whether attenuation has been good or bad. (See data in Laboratory records on Swanswick “acid.”) The trash not only infects the liquor with bacteria but increases aeration. It also seems to carry on a strong infection when fresh juice is added. No marked film forms on the acid so that the acetic bacteria do not have a chance to unfold their full oxidizing activities. B. xylinoides, B. xylinum and Saccharobacillus pastorianus have been isolated from ripe acid.

The acetic bacteria form gluconic and acetic acids. The Saccharobacillus form lactic and acetic acids. The fixed acidity preponderates. (See first and second S.E.S. Reports.)

Jelly and Slime forming Bacteria.— Certain species readily form jelly and slime in weakly acid liquors containing cane sugar. Skimmings and cane juice often undergo a viscous fermentation with the production of gas and the skimmings may frequently be drawn out into long threads (ropy skimmings). This condition interferes with the yeast fermentation which is prolonged and incomplete.

The liquor shows the presence of small cocci single, paired and less frequently in chains. The viscous or ropy condition of the liquor is due to the very diffluent cell walls of the bacteria. The acidity produced in cane juice does not exceed 0.3 per cent, a trace of which is volatile. The growth on cane juice agar is very moist and slimy but no slime is produced on ordinary glucose agar. In glucose broth the chains are very marked so that the organism is a true Streptococcus.

Growth is very rapid in cane juice at 27-40 C. A nonslimy variety has also been isolated. The condition is most marked at the start of crop in both cane juice and skimmings and is due to dirt from the cane and the mill. Owing to its slimy capsule the Streptococcus is not destroyed during tempering in the clarifiers. Thorough cleaning of gutters and skimmings boxes and diluting the hot skimmings with cold water have successfully checked this condition.

Rice Grain.—This occurs not infrequently in washes on estates where no acid is employed. The wash becomes almost filled with gelatinous spherical grains about 1 mm. to 2 m.m. in diameter. The fermentation by the yeast is prolonged and often incomplete. It is caused by a rather thick rodshaped bacterium 1.5-3m by 1 m. Three varieties of this organism have been isolated (see Laboratory records). It strongly resembles the Bacterium vermiforme of Marshall Ward (ginger beer plant). In cane juice the acidity does not exceed .2 per cent. It grows rapidly at 30 C. and just as well in cane juice with 6 per cent, of alcohol by volume as in cane juice without alcohol.

It produces no change in litmus milk and grows out into long more or less cocoid chains without gelatinous sheath in glucose broth. It probably enters the wash from the skimmings.

This organism evidently does not injure the yeast by means of its chemical products. Its interference is physical as the gelatinous masses attach themselves to the yeast cells and grow over them excluding the yeast from contact with the liquor and preventing the cells from rising and whirling in the wash, a condition necessary for active fermentation. A similar kind of interference must be attributed to the streptococcus of viscous ropy skimmings or juice.

Termo bacteria, B. subtilis and B. mesentericus vulgatus have been found in cane juice. These motile forms have not been closely investigated.

The ethers varied from 30,653 to 46,030, and consisted almost wholly of acetic ether. As the ether was formed at the expense of alcohol the yields must be regarded as satisfactory particularly that for the molasses wash to which ammonium sulphate was added. The acid of dunder appears to have an injurious effect on this yeast. Chemical esterification in such liquors could not account for the enormous amount of ether produced. It is evident that both alcohol and acetic acid are formed in the yeast cells by the enzymes, zymase, and oxydase (the latter probably the same enzyme as the oxydase of the acetic bacteria) and are at once brought into union by another enzyme, the whole process occurring within the cell. Certain acetic bacteria are known to yield a vinegar containing a marked amount of acetic ether while other species are quite unable to do so. A yeast is also known which oxydises alcohol to acetic acid and some nonfermenting mycodermas are capable of producing acetic ether in alcoholic liquors. The cells of some species contain, therefore, only the oxydase, others both oxydase and ether producing ferment (esterase). Some of the mycodermas and acetic bacteria which form films on dead washes in Jamaica distilleries may esterify in the way indicated although such forms have not as yet, been isolated in the Laboratory.

Experiments With Fission Yeasts In Dunder And Concentrated Cane Juice Wash.

The juice was boiled down to the consistency of thick syrup without any tempering. The dunder was derived from cane juice and dunder washes from which all alcohol had not been distilled out. This dunder, therefore, had undergone some souring, and was rather high in volatile acidity.—

The yeasts were first grown in a mixture of the cane juice and dunder without added water in sterilized flasks containing 1 litre.

Bottom and top fission yeasts were employed. The wash died in 6 days with bottom yeasts; the wash died in 7 days with top yeasts.

The yeast sediment was then added to 10 litres in large flasks: —

The bottom yeast washes were again dead in 6 days, and the top yeast washes in 7 days.

In this experiment both bottom and top yeasts gave identical attenuations and yields. In spite of the high volatile acidity (45 per cent, of the total) the washes were rapidly (6 and 7 days) worked down with little residual sugar and with excellent results on attenuation and sugar fermented. The rums were high in ethers. Even when distilled as soon as the wash was dead, the rum contained 971 ethers and when the dead wash was allowed to lie six days they were increased to 1404. In washes high in volatile acidity considerable esterification occurs during actual fermentation, and is again greatly increased when the dead wash is allowed to lie.

Grown in pure culture in sterile wash, the top yeast does not yield a rum of higher ether content than the bottom yeast.

Experiment With Poor Dunder.

Washes in litre flasks were set up with dunder water and muscovado sugar.

Oval and fission yeasts were used; the same amount of yeast was added to the sterile wash (in 1 litre flasks) in cash each. To some flasks 10 c.c. of 10% asparagin solution was added at the outset. The loss in weight day by day was as follows:—

The fermentation was normal active with both oval and  fission yeasts in the presence of asparagin. In the absence of asparagin the yeasts scarcely multiplied and the fermentation was very feeble. When, however, asparagin was added on the third day to the latter cultures multiplication set in, and 48 hours later they were fermenting strongly.

These results show clearly that washes set up with a poor dunder are often deficient in nitrogenous food for the yeast with the result that attenuation is feeble and incomplete. In practice such washes are quickly overrun by bacteria, show rapid increase in acidity and become still more unsuited for a vigorous yeast fermentation. In the experimental distillery washes set up with similar dunder worked and attenuated very feebly; 24 hours after addition of 0.15 per cent, ammonium sulphate (equal to 15 lbs. per 1,000 gallons wash) they began to work vigorously and showed a normal attenuation.

Distillery Experiment.

The following experiment selected from a number of such carried out in the Experimental Distillery in 1908, will show what the Jamaica fission yeast are capable of yielding under well regulated conditions.
The fission yeast was of the bottom fermenting type.

It was first developed in a molasses and dunder sterile wash in a flask containing 12 litres (nearly 3 gallons). The yeast was then added to 10 gallons of similar wash in a keg after sterilizing the latter with superheated steam and cooling it to 86 F. When fermentation was almost completed in the keg the contents were stirred up and the liquor added to 50 gallons fresh wash (not sterilized) in a puncheon. When fermentation has started a further 50 gallons wash was added and fermentation allowed to proceed till the wash was quite dead.

The was dead in from 4 to 5 days.

The wash was divided into two portions; the first was distilled for high and low wines (the retorts also receiving charges of wash). The high and low wines were used in toto to charge the retorts for the second distillation 42 gallons wash being introduced into the still.

Taking 6 gallons of rum for every 1° attenuation per 1000 gallons wash as an ordinary average yield 80.4 gallons rum would have been expected on that basis; the actual yield was however 7 per cent, in excess of that amount and must therefore be regarded as highly satisfactory. The above yields are expressed in imperial gallons. In terms of wine gallons the figures are 1-5th. higher, viz., ordinary average yield 96.5 wine gallons. Actual yield 103.2 wine gallons.

Observations Of Estate Materials.

The following determinations made on materials at the estate and on samples at the Laboratory illustrate the composition of dead washes, dunder, and “acids” produced in some “Common Clean” distilleries where rums are made containing 900—1200 parts of ethers per 100,000 alcohol. On such estates no “flavour” is employed so that the ethers in the rums consist wholly or almost wholly of acetic ether.

The acidity of the ripe “acid” varied from 1.9-2.4 per cent, and the gravity from 3.5-10 Brix.

The “Rice Grain” Bacterium.

This organism with remarkably gelatinized cell walls has been already referred to as causing trouble in “common clean” washes particularly in distilleries where only fresh materials are fermented and no “acid” is made.

A sample of dead wash containing the organism was after appropriate dilution plated out in cane juice peptone agar. After some days a variety of colonies including those of yeasts appeared in the medium. Three different types of more or less gelatinous colonies of bacteria could be distinguished.

1. Of no particular shape, raised into a mass and breaking through the agar by rupturing it. These colonies at the surface were smooth, milky, dull and pasty and easily rubbed into a milky homogeneous suspension in water. Such a suspension showed under the microscope small flat, or irregular spherical gelatinous grams about 18-20 microns in diameter and free from any tendency to coalesce with each other. The flat grains showed a more or less circular outline with alternating deep and shallow depressions as indicated in the figure. Embedded in the jelly near the ends of the arms formed by the main depressions and transverse to the surface were two more highly refracting rods separated by the secondary depressions. The almost spherical grains had a convoluted appearance with a fundamentally similar structure, the rods being also transverse to the surface at the ends of the involved arms. This condition was evidently the final state of development of the grain. The resultant appearance was due to the division of rods with gelatinous cell walls, having the property of gelatinous thickening on one side particularly.

The rods are coloured yellow by aqueous iodine in iodide with darker staining granules; the jelly is hardly tinged. The aniline stains colour the rods intensely but do not affect the jelly. The individual rods vary much in length especially in the spheres where they are elongated into threads exceeding 10 micron. The minimum length is 1.5 microns and the breadth about 1 micron. In cane juice peptone broth the organism increases to a finely granular loose deposit in three days at air temperature but multiplies markedly more rapidly at blood heat. The deposit is easily brought into suspension by snaking whereby the minute flocks (grains) are rendered just visible to the naked eye. The liquor overlying the deposit is practically clear. The liquid culture shows a similar appearance under the microscope as the agar material.

Perfectly free rods are not to be found, hence the clearness of the overlying liquor. The organism grows as rapidly and abundantly in cane juice broth containing 6 per cent, alcohol by volume as in broth free from alcohol. The increase of acidity in the broth (equal to 0.1 per cent, at the beginning) does not exceed 0.25 per cent, and only a trace of this is volatile. The acid formed is probably lactic. Gas production is absent or doubtful. In litmus milk the organism produces no change in fourteen days. In nutrient broth and in glucose broth (containing 0.5 per cent glucose) growth is slow with formation of a flocky white deposit. Under the microscope long chains of short rods (almost coccoid) are visible without gelatinous envelopes. The chains grow out from the rods embedded in the grains of the inoculation material, the individual cells being 1.2-1 .5 microns in diameter.

2. Spheres with rough (facetted) surface, translucent, shining, and gelatinous like the agar. The spheres break through the agar and split it into radiating rents. The masses are at least 1 Millimetre in diameter and as a rule from 1-2 m.m. The whole sphere can be lifted away on the loop of the platinum needle. In water it cannot be reduced to a homogeneous suspension but breaks into fragments of jelly. Under the microscope the fragments of jelly are very irregular. The structure is however, very similar (though greatly more irregular) to that of the grains of No. 1. The rods are either transverse at the ends of gelatinous arms or they may be equally gelatinized on both sides. By the rupture of the jelly, the rods often project freely. The length of the rods is very variable, long threads up to 50 microns being frequent. The diameter like No. 1 is almost 1 micron.

In cane juice peptone broth this organism increases by the formation of large irregular masses of jelly or by a gelatinous deposit difficult to raise and then breaking into lumps of jelly. The liquor is distinctly cloudy which is due to the presence of large numbers of free cells with or without gelatinous capsules. The cells are often paired and also form chains of three to ten or more cells. This form also grows equally freely in juice containing 6 percent, alcohol and behaves quite similarly to No. 1 in litmus milk, nutrient broth and glucose broth. The increase of acidity in cane juice broth is also under .25 per cent, and growth at blood heat likewise very rapid.

Grown in conjunction with a bottom fission yeast, both organisms increase freely and the fermentation is 1-2 days more prolonged than in pure yeast culture. The physical interference of the organism with the yeast has been already referred to.

3. Colonies on the surface, transparent, convex, watery (mucoid, not ropy), entire, round and shining. Where colony has broken to the surface a central gelatinous mass, showing under the microscope similar but less marked gelatinous fragments as No. 2 with rods similarly embedded.

The watery part of the colony under the microscope shows rods, single paired and chained with or without an indistinct gelatinous capsule equally developed all round the cells.

In cane juice broth the organism forms an abundant translucent deposit and the liquor is still more cloudy than with No. 2. When shaken up the liquor becomes opaque due to an abundant homogeneous suspension. The dimensions of the rods are the same as for 1 and 2. Its behaviour in litmus milk, nutrient broth and glucose broth is also quite similar. In cane juice broth with 6 per cent, of alcohol the growth was less rapid than in the broth alone. The increase of acidity in cane juice was also under .25 per cent.

The appearance of the colonies applies also to the cane juice agar slants. On this medium each of the three types shows its characteristic growth. It is evident that the three forms are varieties of the same organism. In No. 1 the development of the grain is much more limited than in No. 2. In No. 3 the jelly is less robust and more diffluent, and may be compared with agar which has lost its property of solidifying by heating in a strongly acid liquor. Reference has already been made to the strong resemblance particularly of variety No. 2 to Bacterium Vermiforme of Marshall Ward.

Orange Wine.

Enquiries from several sources came to the Laboratory in the autumn of 1907 and again in the spring of 1908, as to the best way of making orange wine by direct fermentation of the sweetened juice. No experiments had at that time been carried out in connection with the orange wine making and there appeared to be no literature on the subject. The so-called orange wine on the market appeared to consist of diluted rum flavoured with orange essence (or the essential oil from the rind), and highly sweetened. This was more in the nature of a cordial or liqueur and could not be regarded as in any way a true wine.

A preliminary experiment was therefore started in March and April 1908.

A bottom fermenting fission yeast was selected to carry out the fermentation owing to the known resistant properties of fission yeast in general. In order to acclimatize the yeast to a liquor containing a high proportion of citric acid it was first grown in a mixture of molasses, water, and citric acid; the composition was—

The yeast attenuated this wash in four days to 2 Brix, and while it was still working 100 c.c. was used to start a fresh wash prepared from orange juice.

The orange juice was obtained by squeezing the juice of ripe oranges with the rind entirely removed, through a linen cloth.

Gravity of juice—13.8 Brix.
Acidity of juice—1.08 per cent.

To 1,500 c.c. of this unsterilized must, in which cane sugar had been dissolved was added (as stated above) 100 c.c. of the fermenting wash containing the fission yeast. The gravity fell in 7 days from 23.5 to 0.5 Brix, and the final acidity was 1.18 per cent.

After allowing the greater part of the yeast to settle out, the still very cloudy wine was decanted off and bottled. The bottles were filled almost to the corks which were sealed with paraffin. The bottles stood at air temperature for 8 months during which the wine had become perfectly clear and of a dark sherry colour.

The wine had a pleasant aroma of orange and an agreeable though rather marked acid taste. The palatability of the dry wine was improved by the addition of 10 per cent, pure white cane sugar. After this addition the wine was readily appreciated when drunk alone and was also found to be very refreshing beverage when consumed with the addition of two parts of Soda Water. It was pointed out, however, that this wine was not so strongly flavoured as that made by orange growers, and this was attributed to the fact that the oranges had not been squeezed with the rinds still on. The usual practice was to cut the entire oranges into quarters and squeeze out the juice in a wooden press operated by hand. In this way a part of the essential oil contained in the outer rind was set free and entered the juice. To this juice it was customary to add a small proportion of lemon juice (a sample showed a gravity of 10.4 Brix and an acidity of 3. 25 per cent.) to improve the flavour and increase the acidity. White albion sugar was then dissolved in the juice until the gravity was raised to 22-24 Brix. To get this must fermenting a “starter” was then added.

This was prepared by mixing muscovado sugar with warm water to a gravity of about 15 Brix and allowing this to set up a spontaneous fermentation occasioned by the cells and spores of yeasts contained in the sugar. As soon as this was working strongly it was poured into the orange must. If a successful fermentation was set up in the must the latter worked for one to two weeks and finally stopped before all the sugar was worked out, or was intentionally stopped by decanting off the liquor from the yeast deposit. It was pointed out that this method of fermenting the must had some disadvantages namely:—

1. The “starter” spoiled the natural flavour of the wine owing to the characteristic taste of the sugar used.
2. Fermentation often failed in the must after the addition of the “starter” or the fermentation rapid at first fall away quickly and left a product containing insufficient alcohol and too much sugar. This cleared badly and often turned sour (vinegar).

A pure culture yeast acclimatized to orange juice at the Laboratory appeared therefore to offer the most promising solution to the problem. The fission yeasts, well suited to acid distillery washes do not give a pleasant aroma to fermented must. On the other hand a pastorianus yeast isolated from molasses was found to yield a product of very agreeable aroma. It has therefore been employed in the experiments detailed below. Some preliminary work with this yeast indicated that one or more substances contained in the rind of the orange exercised an injurious effect on it when present in the juice from the outset. Juice was accordingly prepared from the fruit after removing the rind. When fermentation was active the liquor obtained by squeezing the rinds separately was added in order to increase the flavour of the finished product.

The yeast was first grown in a wash of molasses, citric acid and ammonium phosphate, then in a mixture of that wash with increasing amounts of orange juice and finally added to the orange juice must. The juice as squeezed from the fruit showed—
Gravity—11. 85 Brix.
Acidity—1. 12 per cent.

The gravity of this juice was increased to 20 Brix by the addition of white crystal sugar, and 0.1 per cent, ammonium citrate added. To start this must one-tenth its volume of a fermenting juice was added which had been attenuated by the pastorianus yeast from 19 Brix to 8.3 Brix. Two days after fermentation began one-sixth of its volume of liquor squeezed from the rinds alone was added. The gravity fell from 20 Brix to 0.3 Brix in 11 days and the must was then dead. After the bulk of the yeast had settled the wine was bottled and kept at air temperature for 15 months. An examination of the perfectly clear dark sherry coloured wine after that period yielded the following figures:—
Acidity—0.64 per cent.
Sugar—0.60
Alcohol as proof spirit—21.43 per cent.

The wine had a fine sherry like aroma and was very palatable after the addition of 10 per cent, cane sugar.

Another must fermented by the same yeast a week later was set up from a juice of—
Gravity—12.05 Brix.
Acidity—1. 18 per cent.

Sugar was added to raise the gravity to 20.8 Brix. This must underwent a more prolonged fermentation and ceased to work with an appreciable amount of sugar unfermented. The must attenuated from 20.8 Brix to 2.8 Brix in 24 days. It was then bottled and cleared very slowly. Fifteen months later the perfectly clear wine showed:—
Total acidity—1.10 per cent.
Volatile acidity—0.15 per cent.
Alcohol as P.S.—18.57 per cent.
In aroma and taste it scarcely differed from the other wine.

When fresh juice is allowed to ferment spontaneously it works slowly and finally dies before all the sugar is fermented. A white dry film usually forms on the surface consisting of mycoderma or a species of Monilia while Apiculatus yeast is often abundant in the deposit. The Apiculatus yeast can only ferment the invert sugar and leaves the cane sugar untouched. As the juice contains about half the total sugar as cane sugar the attenuation stops half way.

A juice worked for 6 days and attenuated from 11.85 to 6.45 and went no further. In another portion of the juice a little added fission yeast reduced the gravity from 11.70 to 1.75 Brix owing to the power of inverting the cane sugar.

When used in larger bulks (10-15 gallons) of sweetened orange juice prepared by pressing the oranges with the rind on, the pastorianus yeast has several times failed to yield a satisfactory fermentation, results which raised the question as to whether this yeast is really well adapted for working in sweetened juice as usually set up. Fermentation certainly sets in more rapidly and vigorously if the sugar is previously melted in hot water to a consistency of syrup and then raised to the boiling point after the addition of 5 per cent. citric acid. This causes the inversion of the bulk of the sugar. The first experiment indicates that the fission yeasts work readily in sweetened juice and it will probably be safer to employ such yeasts in spite of the fact that they do not yield such a good flavoured wine.

The data set out in this paper must be regarded as of the pioneering order and should prove useful as a start in the solution of the difficulties connected with the making of genuine orange wine.

Orange Vinegar.

This is a product with which the wine maker has often hail involuntary acquaintance. About 2½ gallons of an excellent vinegar have been made at the Laboratory in the following way:—
Juice was extracted from the fruit freed from rind.
The gravity was:—10.6 Brix.
Acidity—0.80 per cent.

To this was added sugar inverted by boiling with 2% citric acid. The gravity of the sweetened must was 16.5 Brix. It was pitched with the pastorianus yeast. After 9 days the liquor was dead and showed a gravity of 0.5 Brix. It was allowed to stand in a large flask with a loosely fitting cotton wool plug. After a few weeks an acetic film developed on the liquor and after a further month this had broken up and the liquor was fairly clear.

The total acidity was—5.35 per cent.
Volatile acidity—4.0 per cent.
equal to nearly 5 per cent, of acetic acid. The vinegar was rendered practically clear by filtration through cellulose (filter or blotting paper pulped in water).

Yeast Cultures In Cane Juice Peptone Broth.

Inoculated 20 May, 1910. Frendenreioh flasks.

1. Beer yeast from Jorgensen’s Laboratory, Copenhagen maintained in cane juice broth at Hope. Sets up a speedy fermentation after 12 months in the broth. Used in top fermenting breweries in Denmark. Has been employed successfully in Kingston in a wort of brown sugar, hops and water.
2. American whisky yeast—same source—dextrin fermenting power not tested.
3. Sacchs. thermantitonum—same source—an oval building yeast, with an alleged high optimum temperature for both growth and fermentation. This strain shows nothing striking in those respects.
4. Bottom fermenting oval buckling yeast—the typical yeast of cane juice and washes of relative low acidity. Has been sent to estates in 1908, 09 and 10.
5. Top fission yeast—has been three years in culture. Isolated from a Bluecastle wash. This culture has been used for supplying estates in 1908, 09 and 10.
6. Bottom fission yeast—three years in culture; originally from Mesopotamia wash. A typical bottom yeast.
7. Bottom fission yeast with slight top phenomena from Friendship wash—three years old. Has preserved its power of vigorous fermentation better than No. 6. Has been sent to estates as “bottom yeast” in 1909 and 1910.
8. Bottom fission yeasts—isolated in spring of 1910 from a sample of Swanswick “acid.” Its fermenting power not yet tested in 1 litre portions of wash.
9. Bottom fission yeast—from Long Pond “acid” in Spring 1910. Not yet tested.
10. Top fission yeast—from Long Pond wash, 1910. Not tested.
11. Top fission yeast—from Swanswick wash 1910. Not tested.
12. Sacchs. ludwigii—from Long Pond “acid” 1910 apparently top fermenting.
13. Same as 12—from a different plating.
14. Chained budding yeast—fron Swanswick “acid”. Not investigated.
15. Narrow oval budding yeast—from Long Pond “acid.” Not investigated—may be a variety of No. 4.
16. Pastorianus yeast—from molasses—three years in culture. Used in “orange wine” experiments.
17. Budding yeast—from Parnassus and Moneymusk discoloured crystal sugars. Very abundant in sugars. May be a Torula identical with the torula causing foaming of stored molasses. Does not ferment cane sugar.
18. Willia anomalous (“Anomalus” fruit ether yeast) three years in culture from molasses. This culture was used in experiments with the fruit ether yeast.
19. Mycoderma sp.—from Long Pond dead wash.
20. Mycoderma sp. mixed with B xylinoides—from Swanswick “acid” (See remarks on “ester formation”).
21. B. xylinoides—from Long Pond ”acid” About 1% alcohol was added to the broth.
22. B. xylinoides—from Swanswick “acid.” Alcohol added to broth.
23. Oval yeast and Rice grain variety 2.
24. Large oval yeast—three years in culture.

Literature Of “rum” And “fermentation.”

Rum.—Literature very scanty.
In Library:—
Uber Brauntwein .. Eugene Sell.
Articles by P. Greg (Mesopotamia Estate) in Bulletin Botanical
Department, Jamaica.
—Vol. 2. pts 3, 8, 0.
—Vol. 3. pt. 1.
Sugar Experiment Station Reports 1 and 2.
Bulletin Dept. Agr. (new series) Vol. 1. Xo. 1.
Bulletin Dept. Agr. (new series) Vol. 1. No. 3.
Regarding “artificial rum” see
—Rum Arrack etc., by Gaber.
—Report Whisky Commission 1908.
Fermentation —