Chapter XXXIV: The Evolution of Coffee Apparatus (1)

_Showing the development of coffee-roasting, coffee-grinding,
coffee-making, and coffee-serving devices from the earliest time to
the present day--The original coffee grinder, the first coffee
roaster, and the first coffee pot--The original French drip pot,
the De Belloy percolator--Count Rumford's improvement--How the
commercial coffee roaster was developed--The evolution of
filtration devices--The old Carter "pull-out" roaster--Trade
customs in New York and St. Louis in the sixties and seventies--The
story of the evolution of the Burns roaster--How the gas roaster
was developed in France, Great Britain, and the United States_

A book could be written on the subject of this chapter. We shall have to be content to touch briefly upon the important developments in the devices employed. The changes that have taken place in the preparation of the drink itself will be discussed in chapter XXXVI.

In the beginning, that is, in Ethiopia, about 800 A.D., coffee was looked upon as a food. The whole ripe berries, beans and hulls, were crushed, and molded into food balls held in shape with fat. Later, the dried berries were so treated. So the primitive stone mortar and pestle were the original coffee grinder.

The dried hulls and the green beans were first roasted, some time between 1200 and 1300, in crude burnt clay dishes or in stone vessels, over open fires. These were the original roasting utensils.

Next, the coffee beans were ground between little mill-stones, one turning above the other. Then came the mill used by the Greeks and Romans for grain. This mill consisted of two conical mill stones, one hollow and fitted over the other, specimens of which have been found in Pompeii. The idea is the same as that employed in the most modern metal grinder.

Between 1400 and 1500, individual earthenware and metal coffee-roasting plates appeared. These were circular, from four to six inches in diameter, about 1/16 inch thick, slightly concave and pierced with small holes, something like the modern kitchen skimmer. They were used in Turkey and Persia for roasting a few beans at a time over braziers (open pans, or basins, for holding live coals). The braziers were usually mounted on feet and richly ornamented.

About the same time we notice the first appearance of the familiar Turkish pocket cylinder coffee mill and the original Turkish _ibrik_, or coffee boiler, made of metal. Little drinking cups of Chinese porcelain completed the service.

The original coffee boiler was not unlike the English ale mug with no cover, smaller at the top than at the bottom, fitted with a grooved lip for pouring, and a long straight handle. They were made of brass, and in sizes to hold from one to six tiny cupfuls. A later improvement was of the ewer design, with bulbous body, collar top, and cover.

The Turkish coffee grinder seems to have suggested the individual cylinder roaster which later (1650) became common, and from which developed the huge modern cylinder commercial roasting machines.

Ancient Egyptian mortar and pestle, probably used for pounding coffee]

The individual coffee service of early civilization first employed crude clay bowls or dishes for drinking; but as early as 1350, Persian, Egyptian, and Turkish ewers, made of pottery, were used for serving. In the seventeenth century, ewers of similar pattern, but made of metal, were the favorite coffee-serving devices in oriental countries and in western Europe.

Between 1428 and 1448, a spice grinder standing on four legs was invented; and this was later used for grinding coffee. The drawer to receive the ground coffee was added in the eighteenth century.

Between 1500 and 1600, shallow iron dippers with long handles and foot-rests, designed to stand in open fires, were used in Bagdad, and by the Arabs in Mesopotamia, for roasting coffee. These roasters had handles about thirty-four inches long, and the bowls were eight inches in diameter. They were accompanied by a metal stirrer (spatula) for turning the beans.

Also used for grinding coffee]

Another type of roaster was developed about 1600. It was in the shape of an iron spider on legs, and was designed, like that just described, to sit in open fires. At this period pewter serving pots were first used.

Between 1600 and 1632, mortars and pestles of wood, iron, brass, and bronze came into common use in Europe for braying the roasted beans. For several centuries, coffee connoisseurs held that pounding the beans in a mortar was superior to grinding in the most efficient mill. Peregrine White's parents brought to America on the _Mayflower_, in 1620, a wooden mortar and pestle that were used for braying coffee to make coffee "powder."

When La Roque speaks of his father bringing back to Marseilles from Constantinople in 1644 the instruments for making coffee, he undoubtedly refers to the individual devices which at that time in the Orient included the roaster plate, the cylinder grinder, the small long-handled boiler, and _fenjeyns_ (findjans), the little porcelain drinking cups.

When Bernier visited Grand Cairo about the middle of the seventeenth century, in all the city's thousand-odd coffee houses he found but two persons who understood the art of roasting the bean.

About 1650, there was developed the individual cylinder coffee roaster made of metal, usually tin plate or tinned copper, suggested by the original Turkish pocket grinder. This was designed for use over open fires in braziers. There appeared about this time also a combined making-and-serving metal pot which was undoubtedly the original of the common type of pot that we know today.

There appeared in England about 1660, Elford's white iron machine (sheet iron coated with tin) which was "turned on a spit by a jack.[362]" This was simply a larger size of the individual cylinder roaster, and was designed for family or commercial use. Modifications were developed by the French and Dutch. In the seventeenth century the Italians produced some beautiful designs in wrought-iron coffee roasters.

1--Bagdad coffee-roasting pan and stirrer. 2--Iron mortar and pestle used for pounding coffee. 3--Coffee mill used by General and Mrs. Washington. 4--Coffee-roasting pan used at Mt. Vernon. 5--Bagdad coffee pot with crow-bill spout]

Before the advent of the Elford machine, and indeed, for two centuries thereafter, it was the common practise in the home to roast coffee in uncovered earthenware tart dishes, old pudding pans, and fry pans. Before the time of the modern kitchen stove, it was usually done over charcoal fires without flame.

The improved Turkish combination coffee grinder with folding handle and cup receptacle for the beans, used for grinding, boiling, and drinking, was first made in Damascus in 1665. About this period, the Turkish coffee set, including the long-handled boiler and the porcelain drinking cups in brass holders, also came into vogue.

In 1665, Nicholas Book, "living at the Sign of the Frying Pan in St. Tulies street," London, advertised that he was "the only known man for making of mills for grinding of coffee powder, which mills are sold by him from forty to forty-five shillings the mill."

By combining the long-handle idea contained in the Bagdad roaster with that of the original cylinder roaster, the Dutch perfected a small, closed, sheet-iron cylinder-roaster with a long handle that permitted its being held and turned in open fire places. From 1670, and well into the middle of the nineteenth century, this type of family roaster enjoyed great favor in Holland, France, England, and the United States, more especially in the country districts. The museums of Europe and the United States contain many specimens. The iron cylinder measured about five inches in diameter, and was from six to eight inches long, being attached to a three or four foot iron rod provided with a wooden handle. The green coffee was put into the cylinder through a sliding door. Balancing the roaster over the blaze by resting the end of the iron rod projecting from the far end of the roasting cylinder in a hook of the usual fireplace crane, the housekeeper was wont slowly to revolve the cylinder until the beans had turned the proper color.

A fine specimen in the Peter collection, United States National Museum]

Portable coffee-making outfits to fit the pocket were much in vogue in France in 1691. These included a roaster, a grinder, a lamp, the oil, cups, saucers, spoons, coffee, and sugar. The roaster was first made of tin plate or tinned copper; but for the aristocracy silver and gold were used. In 1754, a white-silver coffee roaster eight inches long and four inches in diameter was mentioned among the deliveries made to the army of the king at Versailles.

Left, seventeenth-century coffee grinder in the Musée de la Porte de Hal--Center, wall mill, eighteenth century--Right, iron mill, eighteenth century]

Humphrey Broadbent, "the London coffee man" wrote in 1722:

I hold it best to roast coffee berries in an iron vessel full of
little holes, made to turn on a spit over a charcoal fire, keeping
them continually turning, and sometimes shaking them that they do
not burn, and when they are taken out of the vessel, spread 'em on
some tin or iron plate 'till the vehemency of the heat is vanished;
I would recommend to every family to roast their own coffee, for
then they will be almost secure from having any damaged berries, or
any art to increase the weight, which is very injurious to the
drinkers of coffee. Most persons of distinction in Holland roast
their own berries.

Left, bronze (Germany)--Center, brass (England)--Right, bronze (Holland, 1632)]

Between 1700 and 1800, there was developed a type of small portable household stove to burn coke or charcoal, made of iron and fitted with horizontal revolving cylinders for coffee roasting. These were provided with iron handles for turning. A modification of this type of roaster under a three-sided hood, and standing on three legs, was designed to sit on the hearth of open fireplaces, close to the fire or in the smoldering ashes. Because of its greater capacity, it was probably used in the inns and coffee houses for roasting large batches. Still another type, which made its appearance late in the eighteenth century, was the sheet-iron roaster suspended at the top of a tall, iron, box-like compartment, or stove, in which the fire was built. This, too, was designed to roast coffee in comparatively large quantities. In some examples it was provided with legs.

Great silver coffee pots ("with all the utensils belonging to them of the same metal") were first used by Pascal at St.-Germain's fair in Paris in 1672. It remained for the English and American silversmiths to produce the most beautiful forms of silver coffee pots; and there are some notable collections of these in England and the United States.

The oriental serving pot was nearly always of metal, tall, and, in old models, of graceful curve, with a slightly twisted ornamental beak in the form of an S, attached below the middle of the vessel. A handle ornamented in the same way formed a decorative balance.

In 1692, the lantern straight-line coffee serving pot with true cone lid, thumb-piece, and handle fixed at right angle to the spout, was introduced into England, succeeding the curved oriental serving pot. In 1700, coffee pots made of cheaper metals, like tin and Britannia ware, began to appear on the home tables of the people. In 1701, silver coffee pots appeared in England having perfect domes and bodies less tapering. Between 1700 and 1800, silver, gold, and delicate porcelain serving pots were the vogue among European royalty.

Both the cast-iron spiders and the long-handled roasters were used in open fireplaces previous to 1770]

In 1704, Bull's machine for roasting coffee was patented in England. This probably marks the first use of coal for commercial roasting.

In 1710, the popular coffee roaster in French homes was a dish of varnished earthenware. This same year a novelty was introduced in France in the shape of a fustian (linen) bag for infusing ground coffee.

By 1714, the thumb-piece on English serving pots had disappeared, and the handle was no longer set at a right angle to the spout. English coffee-pot bodies showed a further modification in 1725, the taper becoming less and less.

Coffee grinders were so common in France in 1720 that they were to be had for a dollar and twenty cents each. Their development by the French had been rapid from the original spice grinder. At first, they were known as coffee mills; but in the eighteenth century, roasters came to be known by that name. They were made of iron, retaining the same principle of the horizontal mill-stones--one of which is fixed while the other moves--that the ancients employed for grinding wheat. They were squat, box-shaped affairs, having in the center a shank of iron that revolved upon a fixed, corrugated iron plate. There was also the style that fastened to the wall. At first, the drawer to receive ground coffee was missing, but this was supplied in later types. Before its invention, the ground coffee was received in a sack of greased leather, or in one treated on the outside with beeswax--probably the original of the duplex paper bag for conserving the flavor.

It succeeded the cast-iron spider, and was suspended from a crane, or stood in the embers]

Early seventeenth century, as pictured by Dufour]

The French brought their innate artistic talents to bear upon coffee grinders, just as they did upon roasters and serving pots. In many instances they made the outer parts of silver and of gold.

By 1750, the straight-line serving pot in England had begun to yield to the reactionary movement in art favoring bulbous bodies and serpentine spouts.

About 1760, French inventors began to devote themselves to improvements in coffee-making devices. Donmartin, a Paris tinsmith, in 1763, invented an urn pot that employed a flannel sack for infusing. Another infusion device, produced the same year by L'Ainé, also a tinsmith of Paris, was known as a _diligence_.

A complete revolution in the style of English serving pots took place in 1770, with a return to the flowing lines of the Turkish ewer; and between 1800 and 1900, there was a gradual return to the style of serving pot having the handle at a right angle to the spout.

Nineteenth century]

In 1779, Richard Dearman was granted an English patent on a new method of making mills for grinding coffee. In 1798, the first American patent on an improved coffee grinding mill was granted to Thomas Bruff, Sr. It was a wall mill, fitted with iron plates, in which the coffee was ground between two circular nuts, three inches broad and having coarse teeth around their centers and fine shallow teeth at the edges.

De Belloy's (or Du Belloy's) coffee pot appeared in Paris about 1800. It was first made of tin; but later, of porcelain and silver--the original French drip pot. This device was never patented; but it appears to have furnished the inspiration for many inventors in France, England, and the United States. The first French patent on a coffee maker was granted to Denobe, Henrion, and Rouch in 1802. It was for a "pharmacological-chemical coffee-making device by infusion." Charles Wyatt obtained a patent the same year in London on an apparatus for distilling coffee. The De Belloy pot is illustrated on page 622.

In 1806, Hadrot was granted a French patent on a device "for filtering coffee without boiling and bathed in air." This use of the word filtering was misleading, as it was many times after in French, English, and American patent nomenclature, where it often meant percolation or something quite different from filtration. True percolation means to drip through fine interstices of china or metal. Filtration means to drip through a porous substance, usually cloth or paper. De Belloy's pot was a percolator. So was Hadrot's. The improvement on which Hadrot got his patent was to "replace the white iron filter (sic) used in ordinary filtering pots by a filter composed of hard tin and bismuth" and to use "a rammer of the same metal, pierced with holes." The rammer was designed to press down and to smooth out the powdered coffee in an even and uniform fashion. "It also," says Hadrot in his specification, "stops the derangement which boiling water poured from a height can produce. It is held by its stem a half inch from the surface of the powder so that it receives only the action of the water which it divides and facilitates thus the extraction which it must produce in each of the particles."

Essex Institute, Salem, Mass.]

A coffee percolator was invented in Paris about 1806 by Benjamin Thompson, F.R.S., an American-British scientist, philanthropist, and administrator. He was known as Count Rumford, a title bestowed on him by the Pope. Rumford's invention was first given to the public in London in 1812. He has gained great credit for his device, because of an elaborate essay that he wrote on it in Paris under the title of _The excellent qualities of coffee and the art of making it in the highest perfection_, and that he caused to be published in London in 1812. It was a simple percolator pot provided with a hot-water jacket, and was a real improvement on the French drip or percolator coffee pot invented by De Belloy, but not at all unlike Hadrot's patented device. Count Rumford, however, was a picturesque character, and a good advertiser. He is generally credited with the invention of the coffee percolator; but examination of his device shows that, strictly speaking, the De Belloy pot was just as much a percolator, and apparently antedated it by about six years.

_Cafetière à la_ De Belloy]

De Belloy employed the principle of having the boiling water drip through the ground coffee when held in suspension by a perforated metal or porcelain grid. This is true percolation. Hadrot did the same thing with the improvements noted above. Count Rumford in his essay admits that this method of making coffee was not new, but claims his improvement was. This was to provide a rammer for compressing the ground coffee in the upper or percolating device into a definite thickness, this being accomplished by providing the perforated circular tin disk water-spreader that rested on the ground coffee with four projections, or feet, that kept the spreader within half an inch of the grid holding the powder in suspension and free from "agitation."

His argument was that two-thirds of an inch of ground coffee should be leveled and compressed into a half-inch thickness before the boiling water was introduced. Practically the same result was achieved in the De Belloy and Hadrot pots, also provided with water-spreaders and pluggers, but the same mathematical exactitude in the matter of the depth of the ground coffee before the percolation started was not assured. De Belloy's spreader did not have the projections on the under side upon which Count Rumford laid such stress. Then there was the hot-water jacket, which was an improvement on Hadrot's hot air bath. Inventors that followed Rumford have made light of the importance that he attached to scientific accuracy in coffee-making; but it is interesting to note how many of the features of the De Belloy, Hadrot, and Rumford pots have been retained in the modern complex coffee machines, and in most of the filtration devices.

These are in the Metropolitan Museum and are of nineteenth century design]

French inventors continued to apply themselves to coffee-roasting and coffee-making problems, and many new ideas were evolved. Some of these were improved upon by the Dutch, the Germans, and the Italians; but the best work in the line of improvements that have survived the test of time was done in England and the United States.

In 1815, Sené was granted a French patent on "a device to make coffee without boiling." In 1819, Laurens produced the original of the percolation device in which the boiling water is raised by a tube and sprayed over the ground coffee. The same year Morize, a Paris tinsmith and lamp-maker, followed with a reversible, double drip pot which was the pioneer of all the reversible filtration pots of Europe and America. Gaudet, another tinsmith, in 1820, patented an improvement on the percolator idea, that employed a cloth filter. By 1825, the pumping percolator, working by steam pressure and by partial vacuum, was much used in France, Holland, Germany, and Austria.

Meanwhile, it was common practise to roast coffee in England in "an iron pan or in hollow cylinders made of sheet iron"; while in Italy, the practise was to roast it in glass flasks, which were fitted with loose corks. The flasks were "held over clear fires of burning coals and continually agitated." Anthony Schick was granted an English patent in 1812, on a method, or process, for roasting coffee; but as he never filed his specifications, we shall probably never know what the process was. The custom of the day in England was to pound the roasted beans in a mortar, or to grind them in a French mill.

In 1822, Louis Bernard Rabaut was granted an English patent in which the French drip process was reversed by using steam pressure to force the boiling water upward through the coffee mass. Casseneuve, a Paris tinsmith, seems to have patented practically the same idea in France in 1824. Casseneuve employed a paper filter in his machine.

In America, a United States patent was granted in 1813 to Alexander Duncan Moore of New Haven on a mill "for grinding and pounding coffee." This was followed by a patent granted to Increase Wilson, of New London, in 1818, on a steel mill for grinding coffee.

Left to right, they are German, Flemish, English, and Dutch specimens in the Metropolitan Museum]

Left, drip pot of 1806--Next two, Durant's inner-tube pot, 1827--Next (fourth), Gandais' first practicable percolator, 1827--Right, Grandin & Crepeaux' percolator, 1832]

In 1815, Archibald Kenrich was granted a patent in England on "mills for grinding coffee."

The coffee biggin, said to have been invented by a Mr. Biggin, came into common use in England for making coffee about 1817. It was usually an earthenware pot. At first it had in the upper part a metal strainer like the French drip pots. Suspended from the rim in later models there was a flannel or muslin bag to hold the ground coffee, through which the boiling water was poured, the bag serving as a filter. The idea was an adaptation of the French fustian infusion bag of 1711, and of other early French drip and filtration devices, and it attained great popularity. Any coffee pot with such a bag fitted into its mouth came to be spoken of as a coffee biggin. Later, there was evolved the metal pot with a wire strainer substituted for the cloth bag. The coffee biggin still retains its popularity in England.

Left, Casseneuve's filter-paper machine, 1824--Center, Gaudet's cloth-filter pot, 1820--Right, Raparlier's percolator]

While French inventors were busy with coffee makers, English and American inventors were studying means to improve the roasting of the beans. Peregrine Williamson, of Baltimore, was granted the first patent in the United States for an improvement on a coffee roaster in 1820. In 1824, Richard Evans was granted a patent in England for a commercial method of roasting coffee, comprising a cylindrical sheet-iron roaster fitted with improved flanges for mixing; a hollow tube and trier for sampling coffee while roasting; and a means for turning the roaster completely over to empty it.

The next year, 1825, the first coffee-pot patent in the United States was granted to Lewis Martelley of New York. It marked the first American attempt to perfect an arrangement to condense the steam and the essential oils and to return them to the infusion. In 1838, Antoni Bencini, of Milton, N.C., was granted a similar patent in the United States. Rowland, in 1844, and Waite and Sener, in their Old Dominion pot of 1856, tried for the same result, namely, the condensation of the steam in upper chambers.

Left, Waite & Sener's Old Dominion pot--Right, Bencini's steam condenser]

The French meantime focused on coffee makers; and in 1827, Jacques Augustin Gandais, a manufacturer of plated jewelry in Paris, produced a really practicable pumping percolator. This machine had the ascending steam tube on the exterior. The same year, 1827, Nicholas Felix Durant, a manufacturer in Chalons-sur-Marne, was granted a French patent on a percolator employing for the first time an inner tube for spraying the boiling water over the ground coffee.

In 1828, Charles Parker, of Meriden, Conn., began work on the original Parker coffee mill, which later was to bring him fame and fortune.

The next year, 1829, the first French patent on a coffee mill was issued to Colaux & Cie. of Molsheim.

That same year, 1829, the Établissements Lauzaune, Paris, began to make hand-turned iron-cylinder coffee-roasting machines.

In 1831, David Selden was granted a patent in England for a coffee-grinding mill having cones of cast-iron.

The first Parker coffee-grinder patent for a household coffee and spice mill was issued in the United States in 1832 to Edmund Parker and Herman M. White of Meriden, Conn. The Charles Parker Company's business was founded the same year. In 1832 and 1833, United States patents were issued to Ammi Clark, of Berlin, Conn., also on improved coffee and spice mills for home use.

Amos Ransom, Hartford, Conn., was granted a United States patent on a coffee roaster in 1833.

The English began exporting coffee-roasting and coffee-grinding machinery to the United States in 1833-34.

1, 2--Improved French drip pots. 3--Persian design. 4--De Belloy pot. 5--Russian reversible pot. 6--New filter machine. 7--Glass filter pot. 8--Syphon machine. 9--Vienna Incomparable. 10--Double glass "balloon" device]

Fig. 1--End elevation. Fig. 2--Front sectional view. Fig. 3--Front elevation, showing how the roasting cylinder was turned completely over to empty. Fig. 4--The examiner, or trier. Fig. 5--Tube (J) to be inserted in H of Fig. 6 to prevent escape of aroma]

It was not until 1836 that the first French patent was issued on a combined coffee-roaster-and-grinder to François Réné Lacoux of Paris. The roaster was made of porcelain, because the inventor believed that metal imparted a bad taste to the beans while roasting.

1--Delephine's coke machine. 2--Bernard's machine, 1841. 3--Circlet for same. 4--Postulart's gas machine]

In 1839, James Vardy and Moritz Platow were granted an English patent on a kind of urn percolator employing the vacuum process of coffee making, the upper vessel being made of glass. The first French patent on a glass coffee-making device, using the same principle, was granted to Madame Vassieux, of Lyons, in 1842. These were the forerunners of the double glass "balloons" for making coffee which later on, in the early part of the twentieth century, attained much vogue in the United States. They were very popular in Europe until the latter part of the nineteenth century.

In 1839, John Rittenhouse, of Philadelphia, was granted a United States patent on a cast-iron mill designed to handle the problem of nails and stones in grinding coffee. His improvement was intended to prevent injury to the grinding teeth by stopping the machine.

In 1840, Abel Stillman, Poland, N.Y., was granted a United States patent on a family coffee roaster having a mica window to enable the operator to observe the coffee while roasting. (See 10, page 630.)

In 1841, William Ward Andrews was granted an English patent on an improved coffee pot employing a pump to force the boiling water upward through the coffee, which was contained in a perforated cylinder screwed to the bottom of the pot. This was Rabaut's idea of nineteen years before. We find it again repeated in the United States in a machine which appeared on the New York market in 1906.

In 1841, Claude Marie Victor Bernard, of Paris, was granted a French patent on a coffee roaster, which was an improvement designed to bring the roasting cylinder and the fire in closer contact. This was accomplished, to quote the quaint language of the inventor, by applying movable legs and "by superimposing a sheet iron circlet around the edge of the furnace to get double the quantity of heat and it presents so much advantage that it has seemed to me worthy of being patented." (See 4, page 627.)

But the French were only toying with the roaster, because roasting in France was not yet a separate branch of business, as it had become in England and the United States, where keen minds were already at work on the purely commercial coffee-roasting machine. The application of intensive thought in this direction was destined to bear fruit in America in 1846, and in England in 1847.

French inventive genius continued to occupy itself with coffee making, and in the invention of Edward Loysel de Santais, of Paris, in 1843, produced the first of the ideas that were later incorporated in the hydrostatic percolator for making "two thousand cups of coffee an hour"[363] at the exposition of 1855, and that has since been improved upon by the Italians in their rapid-filter machines. It should be noted that Loysel's 2,000 cups were probably demi-tasses. The modern Italian rapid-filter machine produces about 1,000 large coffee cups per hour.

James W. Carter, of Boston, was granted a United States patent in 1846 on his "pull-out" roaster; and this was the machine most generally employed for trade roasting in America for the next twenty years. Carter did not claim to have invented the combination of cylindrical roaster and furnace; but he did claim priority for the combination, with the furnace and roasting vessel, of the air space, or chamber, surrounding it, "the same being for the purpose of preventing the too rapid escape of heat from the furnace when the air chamber's induction and eduction air openings or passages are closed."

The Carter "pull-out," was so called because the roasting cylinder of sheet iron was pulled out from the furnace on a shaft supported by standards, to be emptied or to be refilled from sliding doors in its "sides." It was in use for many years in such old-time plants as that of Dwinell-Wright Company, 25 Haverhill Street. Boston; by James H. Forbes and William Schotten in St. Louis; and by D.Y. Harrison in Cincinnati.

The picture of a roasting room with Carter machines in operation, reproduced here, recalled to George S. Wright, the present head of the Dwinell-Wright Company's business, the scene as he saw it so many times when, as a boy of ten or twelve, he occasionally spent a day in his father's factory. "The only difference I notice," he wrote the author, "is that, according to my recollection, there was no cooler box to receive the roasted coffee, which was dumped on the floor where it was spread out three or four inches deep with iron rakes and sprinkled with a watering pot. The contact of water and hot coffee caused so much steam that the roasting room was in a dense fog for several minutes after each batch of coffee was drawn from the fire."

A.E. Forbes also thus recalled the Carter machine in his father's factory in St. Louis in 1853, when he used to help after school; and sometimes ran the roasters, after 1857:

It was barrel shaped, having a slide the full length of one side to
fill and empty. A heavy shaft ran through the centre, resting on
the wall of the furnace at the rear end and on an upright about
eight feet from the front wall. The fire was about sixteen to
eighteen inches below the cylinder and of soft coal. The cylinder
was not perforated, the theory being to keep the vapors from
escaping.[364] This of course was erroneous. The color of the smoke
bursting from the edge of the slide was our medium of telling when
the roasting process was nearing completion, and often the cylinder
was pulled out and opened for inspection several times before that
point was reached. When just right, the belt was shifted to a loose
pulley, stopping the cylinder, which, was pulled off the fire. A
handle was attached to the shaft, the slide drawn, and the coffee
was dumped into a wooden tray which had to be shoved under the
cylinder. The coffee was stirred around in the tray until cool
enough to sack.

The roaster man had to be a husky in those days to pick up a sack
of Rio weighing about one hundred, sixty to one hundred,
seventy-five pounds (not a hundred, thirty-two pounds, as now) and
to empty it in the cylinder. We had no overhead hoppers.

1, 2--English charcoal machines. 3, 5, 8--American coal-stove
roasters. 4--Remington's wheel-of-buckets (American) roaster, 1841.
6--Wood's roaster. 7--Hyde's stove roaster. 9--Reversible stove
roaster. 10--Abel Stillman's stove roaster]

Later we built in the rear and put in two cylinders of the Chris
Abele type, having stationary fronts and filling and emptying from
the front end. We still used soft coal, with the fire sixteen to
eighteen inches under the cylinder.

We had other machines made locally from the Carter pattern. The
idea of the tight cylinder was to keep out smoke, as well as to
keep in the aroma. I think we were the first to use perforations,
because I remember old Jabez Burns coming along after we put in one
of his machines and remarking on it.... We had a kind of mechanical
genius for engineer at that time (he also did the roasting) and he
conceived the idea that we ought to get rid of the moisture in the
roasting coffee because it would cook quicker. When the holes
clogged up, he put in loose pieces of wire bent at the ends which
shook as the cylinder revolved and kept the holes open. Another
thing, he put a hole in the cylinder head and a stopper with a
string on it so he could get out a few grains at a time to note the
progress of the roasting--but he judged mostly by the smoke.

The cooling box was as I have described it, but later we put in a
perforated false bottom which let out some chaff and small stones.

On our first watering, we pulled out the slide and dashed in a
bucket of water, then closed the slide and let it revolve outside
the furnace. This was hard on the cylinder, so later we used the
sprinkling can and put on water sparingly.

Once we had a party that wanted to put in a soapstone lined
roaster, and another near us named Salzgerber patented a
superheated-steam roaster which was shaped like our modern milk
bottle. This was covered with asbestos and worked on a central
bearing so it could be depressed for emptying and elevated for
filling. It did good work.

Mr. Forbes' recollections of the early days of roasting and selling coffee at retail in St. Louis are so illuminating, and paint so interesting a picture of the period that they are printed here to illustrate the conditions that prevailed generally at the time when the commercial roasting machine of the United States was being developed into the modern type. He says further:

Selling roasted coffee was uphill work, as every one roasted coffee
in the kitchen oven. People were buying, say, at twenty cents. Our
asking twenty-five cents "roasted" called for a lot of explanation
about shrinkage, tight cylinders so the strength and flavor could
not get away, etc.; while, when they roasted a pound in the oven
the flavor scented the whole house, thus losing so much strength to
say nothing of the unevenness of their roasts--part raw, part
roasted, producing an unpleasant taste. An occasional burned roast
at home helped some. They tell of a man who, going out in the back
yard and kicking over a clod by accident, uncovered some burned
coffee. He called to his wife and wanted an explanation. She
acknowledged she had burnt it, and hid it so he would not scold. He
said, "We had better buy it roasted in the future and avoid such
accidents."

We roasted in the cellar. We had an elaborately polished Reed &
Mann engine in one window, two brass hoppered mills in the other,
and our boiler was under the sidewalk. We had a mahogany-top
counter, oil paintings on the wall, and bin fronts of Chinamen,
etc., done by the celebrated artist, Mat Hastings (now dead); so
you see we started right.

The fight we had to introduce roasted coffee was fierce. Our
argument was on the saving of fuel, labor, temper, scorched faces,
and anything we could think of. We talked only three coffees, Rio,
Java, and Mocha. When Santos began to come, it was hard to change
them over from the rank Rio flavor to the more mild Santos. The
latter they claimed did not have the rough taste. They missed it
and longed for the wild tang of the Rio.

We did not import, but bought in New Orleans and from several local
wholesale grocers. No one delivered. Shipments were f.o.b. St.
Louis. Draying and packages were extra. Coffee was not cleaned or
stoned, but was sold as it came from the sack. However, we did not
use any very low grades then. If any one complained of the stones
hurting their mills, we advised them to buy ground coffee, showing
how it kept better ground as it was packed tight, whereas the
roasted was looser and the air could get through it. It was fully a
year or more before we began to sell in quantities to make it
profitable. In roasting for others, we got a cent per pound; and
after awhile, that became so much a business it paid all our
expenses. We were the first to roast coffee by steam power west of
the Mississippi and east of the Rocky Mountains.

The tea department helped us to hold out until coffee got its hold
on the public; for in those days every one used tea and insisted on
having it good. Price was no object. How different now!

Five years later (1862) J. Nevison, an Englishman, drifted into
town and opened at 85 North Fourth Street. He got out a very
bombastic circular which caused us to put out the one I enclose
(illustration, page 436). Then came a party named Childs; and after
him, Hugh Menown, grand-uncle of the present Menown, of Menown &
Gregory; and Mat Hunt; all passed over to the Great Majority. After
the Civil War they multiplied pretty fast, coming and going until
now we have nineteen roasting establishments in the city.

The late Julius J. Schotten also wrote the author as follows concerning the days of the Carter roaster and of the wholesale coffee-roasting business founded by William Schotten in 1862:

In the early days, every wholesale grocer was selling coffee; the
wholesale grocer controlled ninety percent of the trade in the
country. It did not pay the coffee roaster to have men on the road
selling coffee in those days. Such being the case, seventy-five
percent of the roasting done by the coffee roasters was job
roasting, at one cent a pound.

In the beginning there were only two kinds of roasted coffee known
to the trade in this section of the country (St. Louis) and of
course one of these brands was "Rio"--the other; "Java". The former
was a genuine Rio, but the Java was mostly Jamaica coffee.

Roasted coffee then was packed (for city trade) in five and ten
pound packages, and this size package seemed to supply the wants of
the ordinary grocer for a week. Occasionally a twenty-five pound
package, and in a few instances as much as fifty pounds of one
grade was sold at a time.

The class of customers the coffee roasters sold in those days were
the smaller merchants; the larger stores, having their ideas as to
quality, bought their coffees green. As they had very little sale
for the roasted, they would send a half-sack, and sometimes a whole
sack to have it roasted. It took a number of years to induce the
larger grocers, and even the average grocers, to purchase their
coffee already roasted.

Coffees were roasted in the old style, "pull-out" roaster cylinder.
That is to say, it was necessary to stop the roaster and to pull
out the cylinder to sample the coffee in order to know when to take
the coffee off the fire. When the coffee was ready to take off, the
cylinder was pulled out its entire length. It was then turned over
and a slide nine inches wide, running the full length of the
cylinder, was opened and the contents were dumped in the cooling
box. When the coffee reached the cooling box, it took two men with
hoes or wooden shovels to stir and turn it until it was properly
cooled, there being no cooling arrangements then as we have
nowadays.

At that time there were no stoning or separating machines; and as a
bag of the ordinary green Jamaica coffee contained from three to
five pounds of stones and sticks, it was necessary to hand-pick the
coffee after it was roasted.

1--English adaptation of French boiler. 2--English coffee biggin. 3--Improved Rumford percolator. 4--Jones's exterior-tube percolator. 5--Parker's steam-fountain coffee maker. 6--Platow's filterer. 7--Brain's Vacuum, or pneumatic filter. 8--Beart's percolator. 9--American coffee biggin. 10--cloth-bag drip pot. 11--Vienna coffee pot. 12--Le Brun's cafetière. 13--Reversible Potsdam cafetière. 14, 15--Gen. Hutchinson's percolator and urn. 16--Etruscan biggin]

After Carter, the next United States coffee-roaster patent was granted to J.R. Remington, of Baltimore, on a roaster employing a wheel of buckets to move the green coffee beans singly through a charcoal heated trough. It never became a commercial success. (See 4, page 630.)

In 1847-48, William and Elizabeth Dakin were granted patents in England on an apparatus for "cleaning and roasting coffee and for making decoctions." The roaster specification covered a gold, silver, platinum, or alloy-lined roasting cylinder and traversing carriage on an overhead railway to move the roaster in and out of the roasting oven; and the "decoction" specification covered an arrangement for twisting a cloth-bag ground-coffee-container in a coffee biggin, or applied a screw motion to a disk within a perforated cylinder containing the ground coffee, so as to squeeze the liquid out of the grounds after infusion had taken place.

The roaster has survived, but the coffee maker was not so fortunate. The Dakin idea was that coffee was injuriously affected by coming in contact with iron during the roasting process. The roasting cylinder was enclosed in an oven instead of being directly exposed to the furnace heat. The apparatus was provided also with a "taster," or sampler, the first of its kind, to enable the operator to examine the roasting berries without stopping the machine. As will be seen by referring to the picture of the model shown, the apparatus was ingenious and not without considerable merit. Dakin & Co. are still in existence in London, operating a machine very like the original model.

In 1848, Thomas John Knowlys was granted a patent in England on a perforated roasting cylinder coated with enamel.

It is to be noted in passing that this idea of handling the green bean with extreme delicacy, evidently obtained from the French, was never taken seriously in the United States, whose inventors chose to handle it with rough courage.

The first English patent on a coffee grinder was granted to Luke Herbert in 1848.

In 1849, Apoleoni Pierre Preterre, of Havre, was granted an English patent on a coffee roaster mounted on a weighing apparatus to indicate loss of weight in roasting and automatically stop the roasting process. At the same time he secured an English patent on a vacuum percolator, not unlike Durant's of 1827.

In 1849 also, Thomas R. Wood, of Cincinnati, was granted a United States patent on a spherical coffee roaster for use on kitchen stoves. It attained considerable popularity among housewives who preferred to do their own roasting. (See 6, page 630.)

In 1852, Edward Gee secured a patent in England on a coffee roaster fitted with inclined flanges for turning the beans while roasting.

C.W. Van Vliet, of Fishkill Landing, N.Y., was granted a United States patent in 1855 on a household coffee mill employing upper breaking and lower grinding cones. He assigned it to Charles Parker of Meriden, Conn. In 1860-61 several United States patents were granted John and Edmund Parker on coffee grinders for home use.

In 1862, E.J. Hyde, of Philadelphia, was granted a United States patent on a combined coffee-roaster and stove fitted with a crane on which the roasting cylinder was revolved and swung out horizontally for emptying and refilling. This machine proved to be a commercial success. Benedickt Fischer used one in his first roasting plant in New York. It is still being manufactured by the Bramhall Deane Company of New York.

In 1864, Jabez Burns, of New York, was granted a United States patent on the original Burns coffee roaster, the first machine which did not have to be moved away from the fire for discharging the roasted coffee, and one that marked a distinct advance in the manufacture of coffee-roasting apparatus. It was a closed iron cylinder set in brickwork. (See illustration, page 635.)

Jabez Burns had been a student of coffee roasting in New York for twenty years before he produced the machine that was to revolutionize the coffee business of the United States. He had brought with him from England a knowledge of the trade in that country, where he first began his business training by selling Java coffee at fourteen cents and Sumatra at eleven cents to hotels, boarding-houses, and private families.

Up to the time of the Civil War, the contrivances employed for roasting coffee in every case necessitated the removal of the roasting apparatus--whether pan, globe, or cylinder--from the fire. The process of causing coffee to discharge from the end of the roasting cylinder at the pleasure of the operator while the cylinder was still in motion was new; and the double set of flanges to produce this effect, and at the same time, during the process of roasting, to keep the coffee equally distributed from end to end of the cylinder, was new. Some one suggested this last improvement was simply an Archimedean screw placed in a cylinder, but Mr. Burns replied: "It is a double screw, a thing never suggested by the Archimedean screw. It is, in fact, a double right and left augur, one within the other, firmly secured together and also to the shell or cylinder, and when the cylinder revolves the desired result is obtained--the idea being entirely original."

Mr. Burns had watched the development of the coffee business from the time when the preparation of coffee was largely confined to the home, where the approved roasting implements were hot stones, or tiles, iron plates, skillets, and frying pans. Some of these were still in use twenty years after he produced his first machine; and he often said that coffee evenly roasted by such methods was just as good as if done by the best mechanical device ever invented. He also said: "Coffee can be roasted in very simple machinery. Some of the best we ever saw was done in a corn popper. Patent portable roasters are almost as numerous as rat traps or churns."

He early saw the practise of domestic roasting falling into disuse, as it was becoming possible to supply the consumer with roasted coffee for only a trifle more than in the green state, with all the labor and annoyance of roasting done away with--a talking point that John Arbuckle was quick to seize upon in his first Ariosa advertising.

In almost every town of any size there were concerns engaged in the roasting business. Within a few years, Burns machines were placed in all the principal roasting centers. Pupke & Reid in New York; Flint, Evans & Co., and James H. Forbes in St. Louis; Arbuckles & Co., in Pittsburgh; the Weikel & Smith Spice Co. in Philadelphia; Theodore F. Johnson & Co., in Newark; Evans & Walker in Detroit; W. & J.G. Flint in Milwaukee; and Parker & Harrison in Cincinnati, were among his first customers.

It is said that in 1845 there were facilities in and around New York to roast as much coffee as was then consumed in Great Britain. Steam power was being extensively used, and the roasting was done here for a large part of the country. The habit was to buy roasted coffee from the coffee and spice mills by the bag or larger quantity for country consumption; and the grocers and small tea stores, for local consumption, bought from twenty-five pounds upward at a time. This method cheapened the roasting of coffee to half a cent a pound; and then good profits could be made, for everything was cheap in those days. Even at that, it would have been impossible for each tea dealer to have roasted his own coffee for several times the amount, so the practise was generally adhered to all over the country.

Jabez Burns wrote in 1874:

It is preposterous to suppose that household roasting will be
continued long in any part of this country, if coffee properly
prepared can be had. This is demonstrated by the remarkable
advances made in Pittsburgh and other places, where only a few
years ago the sales were chiefly in green coffee. Now the amount
roasted in Pittsburgh alone by those who make a business of it,
exceeds the entire consumption of coffee of any kind in the United
States fifty years ago. It will never pay for small stores to roast
if the large manufactories will do the work well, and if they will
not, small dealers will add proper machinery, and will eventually
become strong competing dealers. By doing the work with proper care
they will not only secure a reputation with large sales for
themselves, but will command the roasting for other parties.

Until the Burns roaster appeared, coffee roasters were usually cylinders that revolved upon an axis; the other devices that were tried were not successful. Jabez Burns thus describes the first roaster he ever saw at Hull, England:

It consisted of a furnace, open at the top, and a perforated
cylinder with a slide door. The axis, or shaft, of the cylinder had
bearings on a frame which passed outside the furnace, while the
cylinder went down into the fire pit, the top of which could be
covered over. In this position it could be turned by means of a
crank on the end of a shaft The only means of testing was by the
escape of the steam or aroma, whichever predominated, passing out
through the perforations at the top; but so expert was the operator
and so quick to detect the aroma, that he seldom had to return the
cylinder to the fire to produce a satisfactory roast. This man
roasted fifty pounds or less in a batch for a number of retail
stores.

Globes, consisting of two hemispheres, made of cast-iron and so
arranged that they opened to fill and discharge, but operated
substantially as above, only with the method of lowering into the
fire changed somewhat, I have seen in use in Scotland in 1840. They
were called French roasters.

In this country a few years ago the use of the long sheet-iron
cylinder was almost universal, varying only in the method of
placing the cylinder over the fire--some sideways on a track,
others endwise, sliding on a long shaft or by turning on a crane,
in either case causing considerable labor and loss of time, which
often resulted in the hands of the inexperienced in more or less
spoiling the batch of coffee.

From his expert knowledge of coffee and coffee-roasting problems, Jabez Burns quickly rose to a commanding position in the industry. He was a trade teacher and a trade builder. He had very definite ideas on roasting. He said:

The object of roasting is not attained until all the moisture
(water of vegetation) is driven off. Roast properly--uniformly and
sufficiently--and you will get all the aroma there is in the bean.
Coffees of various kinds can not be roasted to a uniform color.
Some will be of a light shade when sufficiently roasted while
others will have to be roasted dark to develop the aroma.
Therefore, appearance alone is not a proper test. Aroma-saving
devices have had their day. Coffee is of no use unless the aroma is
fully developed, and the more it is developed by roasting the
better it is. What passes off in the roasting process can not be
saved and is so small that if all of it in the country could be
collected and freed of all foreign matter, it would not weigh an
ounce.

Roast coffee over a slow fire so that it will be an hour before it
has the color of roasted coffee, and, in contrast, produce in
another batch of like quantity the same color in thirty minutes,
and it will be found for all intended purposes, either to grind,
sell or drink, that the latter will be, beyond all comparison, the
best. Coffee should be roasted uniform and as quickly as possible,
only it must not be scorched or spotted, otherwise it will have a
bitter burned taste. If roasted properly it will very considerably
increase its bulk and will be plump, swelled out and crisp; easily
crushed in the hand or between the fingers.

In his _Spice Mill Companion_, published in 1879, Jabez Burns said further in regard to roasting: