English wine gallon

For uses of this gallon in countries other than England since 1826, see wine gallon, U.S. gallon

A unit of capacity in England, customarily from at least the 16th century and by law since 1707, = 231 cubic inches, about 3.7854 liters. chart symbol It differs from the corn or dry gallon, the ale gallon and the beer gallon.

Britain abandoned the wine gallon in 1826 when it adopted imperial measure, but the wine gallon is the basis of the U.S. gallon and many other measures.

The wine gallon is sometimes said to have originated in 1706 during the reign of Queen Anne, and is sometimes called the wine gallon of Queen Anne. It is, however, centuries older. See The Carysfort Committee and the Wine Gallon, 1758, excerpts from the report of an extensive government investigation of its origin.

The earliest evidence for the size of the wine gallon is documentary; the earliest surviving wine gallon standards date from the late 15th century.

English units of capacity were usually defined as the volume occupied by a specified weight of some substance; in the case of liquid capacity, either wine or water. The Tractatus (probably middle of the 13th century) says: “Twelve Ounces make a Pound and Eight Pounds make a Gallon of Wine; and Eight Gallons of Wine make a Bushel of London; which is the Eighth Part of a Quarter.” Since no standards made before the 1400's have survived (they were probably made of wood and have decayed), tracing the history of the wine gallons is partly a process of sorting through weight standards and densities, comparing the results with the earliest surviving standards. (For those who wish to experiment, we provide an Excel spreadsheet.) This sort of analysis suggests that the wine gallon was originally 224 cubic inches, became 233 by the 16th century, and by the 17th century was 231 cubic inches.

The 224-cubic-inch wine gallon

In 1645 and 1647 John Wybard measured the capacity of an old wine gallon standard at the Guildhall in London, and found it to be 224 cubic inches.¹ A century of subsequent tests by other competent persons confirmed this value within a few tenths of a cubic inch.

The wine gallon apparently originated as a unit used by merchants in the early medieval English trade importing wine from France, possibly through the Flemish city of Bruges. It is not surprising then that the weight standard appears to be that of Paris. Water weighing 8 livre of Paris occupies a volume of about 224 cubic inches. The livre was divided into 15 ounces, each weighing about 30.6 grams.

An alternative explanation is that the wine gallon was the volume occupied by 8 tower pounds of wheat, which fits the 12-ounce pound requirement of the Tractatus. This of course supposes that the liquid measures were raised from wheat, which seems not to have been the practice in the British Isles, but may have been introduced by the Normans. The 15th century Anglo-Norman manuscript composicione monete et mensurarum explicitly says the wine gallon was 8 pounds of wheat.

Possibly both definitions were correct, and equivalent, but at different periods.

1. See Wybard1650.pdf, and the excerpts in Sources, below.

The 233-cubic-inch wine gallon

John Reynolds, a colleague of Wybard's, focused on the old standards in the Tower of London rather than the Guildhall. The positions he held show that he was a competent and meticulous gauger: assay-Master at the Goldsmiths' Company and under-assayer at the Mint.

In 1641 Reynolds created an elegant half gallon (a pottle) standard, now in the keeping of the Cambridge University Museum of Archaeology and Anthropology¹. It is inscribed “1641 A wine pottle tryed by John Renalds at the Tower”, so presumably it was based on a standard then in the Tower that has not survived. According to a measurement made in 1994, double this pottle's capacity would make a wine gallon of 233.06 cubic inches.

Further evidence that there really was a wine gallon this size is that it is half of the Scottish gallon defined in the 1426 Assize. That gallon, consisting of 8 pints, was 467 cubic inches.

The best explanation of the origin of this gallon is that of Connor and Simpson, who have suggested² that the 233-cubic-inch wine gallon is the volume occupied by water weighing 9 trade pounds, each pound of 15 avoirdupois ounces. Why “9”, “15” and “avoirdupois”?

Why 15 ounces?

The trade pounds of medieval England, France and Scotland had 15 ounces, not 16. (Trade pounds are those generally used for articles made and sold within the country, in contrast to goods being imported or exported, like wine or wool, for which other weight systems were used.) The 224-cubic-inch wine gallon was defined with a 15-ounce pound, and weights and measures officials are generally very conservative, so it is not surprising that any new definition would also call for 15 ounces.

Why avoirdupois ounces?

in the 15th century avoirdupois weight was replacing X weight before the 16 ounce pound assumed  In the same period, the avoirdupois ounce

Why 9 pounds?

The change from 8 pounds per gallon to 9 pounds reflects the practice of the seller giving a bit more. In time this becomes defined and expected part of the quantity. In England as elsewhere continuing battle between the authorities, whose laws insisted on well-defined stricken measure, and the public, who wanted to get and give a bit extra.

The Tractatus says a bushel is 8 wine gallons, but the bronze bushel standard of Henry VII (about 1497) actually contains 9 gallons, because it is meant to contain, stricken, the same quantity that a 8-gallon measure would contain heaped. If Henry's bushel is divided into 8 parts, the parts are 9-pound gallons.

The bushel's gallons, of course, are dry measure, which in England (the Tractatus not withstanding) was actually always raised from wheat. wine gallon was raised from wheat, which it never was If there were 9 pounds of wheat to the corn gallon, there must be 9 pounds of water to the wine gallon.

1. Artefact (Z 2531) Arch.

2. Connor and Simpson, page 234.

The 231-cubic inch wine gallon

It is probable that the 231-cubic inch capacity arose as the result of choosing whole numbers of inches to simplify constructing measures. Taking pi as 22/7, which was the usual practice at the time, a cylinder 6 inches high and 7 inches in diameter has a volume of 231 cubic inches (6 × 22/7 × 7/2 × 7/2), not significantly different from 233. (This process of rounding off dimensions to simplify constructing measures also occurred in 1696 with the Winchester bushel, and accounts for the size of the U.S. bushel.) The 6- and 7- inch dimensions were codified in the law of 1707, but 231 cubic inches had been the accepted value for at least the previous century.

For several hundred years the Excise collected duty on wine imports by the 231-cubic-inch gallon. Then in 1688 someone told the Commissioners of the Excise that a 224-cubic-inch gallon was the “true wine gallon.” The origins of the unit had been forgotten, but on looking into the matter, the Excise found that the custodian at that time of all legal standards, the Exchequer, had a standard for a 272-cubic-inch gallon but none for one of 231 cubic inches.

Changing to a larger gallon would have caused a large drop in the King's revenues (since the tax was on the gallon), so the Commissioners carefully refrained from revising the standard. But the idea was out. In 1700, a canny wine importer, who had been sued by the government for underpayment of import duty, argued in his defense that there was no legal definition for the gallon the customs agents were using. The Crown had to forfeit the case. In 1706 the government remedied the matter by legalizing the 231-cubic-inch wine gallon.¹

1. 5 Anne c 27, § 17, 1706, which took effect March 1, 1707. Statutes at Large, Vol IV, pages 244-245.



Wybard's chapter (1650) on the standard gallons kept in the Guildhall begins: “It is generally holden by Artists about the City of London, that a Wine-Gallon containeth in its concave Capacity, 231 cubicall or solid inches, or is insensibly different therefrom.”


For the second thing, the content of our English Gallon, which is the measure of all these vessels. This is most commonly received, that a Wine-gallon conteins 231 cubick inches : yet Mr. Wybard pleads very strongly, that it is somewhat lesse, making the Wine-gallon to be 225 inches. But the difference being so small, the errour will not be much; and therefore, till the exact truth be more certainly known, I shall, with the most, follow the first; counting it better to allow rather a little over-measure, than any thing under.

Henry Phillippes.
The Purchasers Pattern. 2nd ed., corrected and enlarged.
London: Printed for R. & W. Leybourn, for T. Pierrepont..., 1654.
Page 218.


But Doctor Wybard in his Tectometry, Page 289, doth suppose the Wine Gallon to contain but 224, or 225 Cubick Inches at the most, and pursuant to this account an Experiment was made by Mr. Richard Walker and Mr. Philip Shales, two General Officers in the Excise. They caused a Vessel to be very exactly made of Brass, in form of a Parallelopipedon, each side of its base was 4 Inches, and its depth 14 Inches; so that its just content was 224 Cubick Inches. This Vessel was produced at Guild-Hall in London (May 25th, 1688) before the Lord-Mayor, the Commissioners of Excise the Reverend Mr. Flamstead Astr. Reg. Mr. Halley, and several other Ingenious Gentlemen, in whose presence Mr. Shales did exactly fill the foresaid Brazen Vessel with clear Water, and very carefully emptied it into the old Standard Wine Gallon kept in Guild-Hall, which did so exactly fill it, that all then present were fully satisfied the Wine Gallon doth contain but 224 Cubick Inches. (This notable Experiment I saw tried.) However, for several Reasons, it was at that time thought convenient to continue the former supposed content of 231 Cubick Inches to be the Wine Gallon and that all computations in Gauging should be made from thence, as above.

John Ward.
The Young Mathematician's Guide. Fourth Edition.
London: Printed for A. Batterworth and F. Fayrham, 1724.
Pages 34-35.


The Exchequer Standard Wine Gallon is dated 1707, and was found to contain 133.4 ounces [of Thames water], answering to 230.9 cubic inches. An experiment of Dr. Wollaston and Mr. Carr, in 1814, gave 230.8, the mean being 230.85; while the measurement of 1758 made it 231.2. A duplicate of this measure, and of the same date, is kept at Guildhall.

Dr. Wollaston and Mr. Carr examined also the three other Wine gallons at Guildhall. The oldest of these seems to be the same that was was measured by Halley and Flamsteed in 1688, and was said to contain 224 cubic inches; its actual capacity is 224.4. The Wine Gallon of 1773, which is in daily use for adjusting other measures, was probably in the first instance a correct copy of the Exchequer Gallon, but has been reduced by a bruise, and by the wear of the brim, to 230.0 cubic inches, having lost 4-5ths of a cubic inch, or one three-hundredth of its whole capacity. The wine gallon of 1798 contains 230.9 cubic inches.

The Excise Wine gallon was found, by a similar experiment, to contain 230.1 cubic inches, having partaken of the progressive deficiency of the Guildhall gallon, from which it was derived.

First Report of Commissioners appointed to consider the Subject of Weights and Measures 7 July 1819.
Parliamentary Papers (HC 565).
Appendix A, page 7.


As regards wines and spirits, the old wine gallon of 231 cubic inches is yet the standard in the Collector's office, though the “Imperial” gallon of Act 6, Geo. IV, cap. 74, which came into use in England, 1st January 1826, is 277.24 cubic inches, and is used in the Commissariat and Medical departments.

Manual of the Administration of the Madras Presidency… Volume II.
Madras: Printed by E. Keys, at the Government Press, 1885.
Page 515.

home | units index | search |  contact drawing of envelope |  contributors | 
help | privacy | terms of use