Talk:Comparison of the imperial and US customary measurement systems

"Imperial" Compared with "U.S. Customary"
This section needs a rewrite. Through an accident of history, there is a reason that U.S. definition of the units of gallon and bushel are different from the U.K. "Imperial" definitions. Note that author picks up 1824 definitions of Imperial gallon and bushel. U.S. declared its independence in 1776. Units in use in U.S. were the ones in use in U.K. at that time.

The metric system (now S.I. units) were originally introduced during the Napoleonic era. The English liked the base 10 concept, but not enough to adopt the metric system, so they redefined their gallon from approx 8.6 pounds of water to define the volume of a gallon to 10 pounds of water defining the "Imperial" gallon. This occured after the American Revolution, so the U.S. did not adopt this change. Thus, it is better to call the U.S. system as "U.S. Customary". There are a lot of references at www.nist.gov under "weights and measures" to help here.

--69.140.130.29 20:29, 24 Feb 2005 (UTC)


 * Today's US units are not the ones used in pre-1824 UK, but a selection of them. There were also efforts after 1776 in the US to harmonize and drop some measures or even create a whole new system of units (based upon the second pendulum), lead by Thomas Jefferson, but with the mere result that there are only two gallons (one wet, one dry) left. After all, they had already decimalised their currency, unlike the Britains. But yes, maybe there's need for a rewrite. Perhaps this article and U.S. customary units should be merged and grouped chronologically. Christoph Päper 21:50, 24 Feb 2005 (UTC)

Long & short hundredweights & tons
I'd written that the terms "long" and "short" were used in the U.S. This is what the long ton & short ton pages seem to suggest. Before reading those pages I'd thought that these were just general names not U.S.-only ones but being a metric lad I wasn't 100% sure (not even 50%) so I edited the page to reflect this. If this is not the case, we've got to chang those pages as well. Jimp 15Jul05


 * They aren't U.S.-only names, though some editors hold that mistaken belief. They are common names used by anyone when they need to distinguish between them.  Gene Nygaard 00:19, 15 July 2005 (UTC)

That's exactly what I'd thought before. So, let's fix those pages up. Jimp 15Jul05


 * Long hundreds go back as far as old english, long count is count in 120, short count in decimals. Gwilt (1867), distinguishes between a "long hundred weight" (ie weight of a long hundred = 120 lbs), vs long hundredweight (112 lbs).--Wendy.krieger (talk) 10:15, 15 February 2011 (UTC)

Equivalence of Imperial fluid ounces and avoirdupois ounces of water
The following statement cannot be exactly correct:


 * From this we see that, in the Imperial system, an avoirdupois ounce of water at 62 °F has a volume of one fluid ounce, because 10 pounds is equivalent to 160 avoirdupois ounces, and 1 Imperial gallon is equivalent to 4 Imperial quarts, or 160 fluid ounces. This convenient relation does not exist in the U.S. system because a U.S. gallon of water at 62 °F weighs about 8 1/3 pounds, or 133 1/3 avoirdupois ounces, and the U.S. gallon is equivalent to 4 × 32, or 128 fluid ounces.

The Imperial gallon is defined to have a volume of exactly 4.54609 dm³, and the international pound is exactly 0.45359237 kg. So for an Imperial gallon of water to weigh exactly ten pounds, the density of water would need to be exactly


 * $$\rho = \frac{10 \times 0.45359237}{\left(\frac{1}{10}\right)^3\times 4.54609} \approx 997.76~\mathrm{kg}/\mathrm{m}^3$$.

This relation can only be true at one temperature for a given pressure (about 22 °C or 72 °F at 1 atm pressure), and this temperature/pressure would need to be specified with infinite precision to make the density exact, since the temperature scales already have their own independent definition!

Assuming that the number of fluid ounces in the Imperial gallon is fixed at exactly 160, this implies that the relation between Imperial fluid ounces and avoirdupois ounces of water should be regarded as an approximation, even though it is close enough for most practical purposes.

Mtford 06:51, 29 March 2006 (UTC)


 * No, all that is needed to make it exact is to declare it so, replacing less precise earlier definitions. The original 1824 definition was 10 pounds of water, weighed in air with brass weights at 62 °F.  Your calculations are based on true mass (as if weighed in a vacuum, rather than in air).   Gene Nygaard 20:10, 16 September 2006 (UTC)

Imperial units started exactly on 14 October 1066 ?
The article introduction notes: "These English systems had developed in England over several centuries since the Battle of Hastings in 1066". I question that the English systems started exactly on 14 October 1066. Does the article need to be modified to something like "These English systems were introduced after the Norman Invasion" ?

--mgaved (talk) 18:37, 20 February 2008 (UTC)--mgaved (talk) 18:37, 20 February 2008 (UTC)

The tower pound, of 5400 grains troy, dates from King Offa in 770. The Drusian foot, of 13.2 inches, 15 of which yields the modern rod, pole or perch, is known before the Norman conquest. There are old english measures, such as sheppel (a dry measure).--Wendy.krieger (talk) 08:08, 28 February 2011 (UTC)

U.S.A.
The United States is one of 3 countries that has not officialy adopted the metric system. I removed the error which said it had. 75.118.187.114 (talk) 15:10, 30 December 2008 (UTC)

The U.S. *has* officially adopted the metric system, along with the U.K. and Myanmar. Since the whole argument is based on a flawed CIA study (US had officially been metric even before it was published, and has permitted metric since 1866). Since this non-fact is ubiquitously plastered all over the internet, does that excuse us quoting it here?Surveyor792 (talk) 22:48, 12 October 2013 (UTC)


 * The US ain't 'officially metric'. We passed some legislation (now nearly a half-century ago) declaring it to be so, but allowed that all transitions would be done voluntarily - thus would be encouraged, but not forced. Those statutes were heavily gutted in the 1990s, and since then (at least) all metrication efforts have more or less halted (i.e., those applications which had already converted to metric use metric, but everything else uses standard and is unlikely to change). Some states have even reversed some of their metrication implementations in the last couple decades. Firejuggler86 (talk) 20:15, 11 March 2021 (UTC)

"Food" tags in table
To my knowledge, all liquids trade in the US, including food, is done in fluid ounces, pints, quarts and gallons based on 1 gal = 231 cu in (note: cups and spoons are not legal for trade, page 7). It's only in the "Nutrition Facts" tables on the labeling, in specifying serving sizes, are the approximations of "1 fl oz = 30 mL," "1 tbsp = 15 mL," "1 tsp = 5 mL," and "1 cup = 240 mL" used. This is why the table on a gallon (231 cu in) of milk says there are about 16 servings of "1 cup" (240 mL) each. See Cup (volume). These same nutrition tables approximate in other areas as well, where a can of "Just 1 Calorie" diet cola is labeled as 0 Calories and two "One-and-a-half Calorie Breath Mints" add up to 5 Calories.

These approximations are only presented to the consumer after the information has been determined more precisely in a laboratory (typically on a massic basis), and by looking at the USDA National Nutrient Database entry for water, you'll note that 1 fl oz is listed with a mass of 29.6 g and a cup of volume 8 fl oz of it is listed as 237 g.

The way the table in this article is currently set up implies that the quantity labels on a gallon of milk are different from those used on a gallon of gasoline, that the US system is more complicated than it actually is, and that these approximations for the purposes of nutrition (where 240 cm3 of corn chips aren't going to kill you markedly faster than 1/2 pt) are legal for purposes of trade and taxation.Guppy313 (talk) 17:22, 14 February 2009 (UTC)


 * "...in specifying serving sizes, are the approximations of "1 fl oz = 30 mL," "1 tbsp = 15 mL," "1 tsp = 5 mL," and "1 cup = 240 mL" used."
 * These are not approximations, but legal definitions of the USC units as defined by the FDA. Part of the reason for this is that all package fills are in metric using machines that can only fill in 5 g or 5 mL increments.  If a label says 1 lb (454 g), the actual fill is 460 g.  Even though they can fill to 455 g, the fill is to 460 g, possibly as a precaution to prevent under fills.  Since 29.6 is not an increment of 5 mL, the actual fill is 30 mL for 1 ounces and 240 mL for 8 ounce. Ametrica (talk) 00:36, 16 October 2022 (UTC)

I'd say, either the entries tagged as (food) should be removed from the table, or it needs an explaination. Now it just sits there in the table, confusing the reader. 213.84.146.74 (talk) 06:35, 27 October 2011 (UTC)

Capitilisation
There's no need for capitilisation of imperial. I've reverted the recent page move. J IM ptalk·cont 00:11, 27 June 2009 (UTC)

Centilitre: an unnecessary complication that could be dropped.
I notice that the article and the tables use the centilitre. When Australia introduced the metric system in the 1970s it was not mentioned. We have millilitres and litres but not decilitres or centilitres. I cannot speak for other countries, but as far as I can see, the centilitre is hardly used and can be dispensed with in this article. This would certainly simplify the material in the table as can be seen below:

This would be clearer and simpler and more in line with the way that the metric system is applied. I would be interested to see what other editors think of this proposal.Michael Glass (talk) 08:37, 13 August 2009 (UTC)
 * I couldn't agree more. J IM ptalk·cont 10:05, 13 August 2009 (UTC)

Good. As no-one else has commented I'll make that change. Michael Glass (talk) 11:35, 14 August 2009 (UTC)

Add pints
Now that we've taken out centiliters, I think it'll be helpful to add pints.

While calculating the pint conversions, I noticed some inconsistencies in the quart equivalents, so I've gone back and re-figured them based on these definitions (exact numbers): 1 US liquid gal. = 231 cu. in.; 1 US dry gal. = 268.8025 cu. in.; and 1 Imperial gal. = 4.54609 L. I haven't checked if there may be errors in the gallon and fluid ounce sections, however. BW95 (talk) 05:58, 17 August 2009 (UTC)


 * Sounds good to me. J IM ptalk·cont 07:10, 17 August 2009 (UTC)


 * I just checked over the entire table and have made more corrections. Will wait another day for other comments and then update the main article. BW95 (talk) 08:23, 17 August 2009 (UTC)
 * Changes are made. BW95 (talk) 13:37, 18 August 2009 (UTC)

Typographic note
In the table, why are some numbers in italic, some in bold, and some in regular font? From what I can figure, italic means approximation, bold means definition, and regular means exact. Can someone confirm this? I think it'll be a good idea to make this explicit in the article so that people aren't left wondering. BW95 (talk) 23:07, 2 September 2009 (UTC)
 * I've removed the italics and added a legend. BW95 (talk) 12:46, 8 October 2009 (UTC)

international yard, survey yard, EXACTLY 2 ppm difference
is that strictly true? CorvetteZ51 (talk) 09:04, 8 October 2009 (UTC)
 * Yes. See the NIST's explanation: "1 international foot = 0.999 998 U.S. survey foot exactly." (http://physics.nist.gov/Pubs/SP811/appenB.html#B.6) A yard is 3 feet in either definitioin. BW95 (talk) 11:55, 8 October 2009 (UTC)

1 metre = 39.37 inches; 1 inch = 2.54 cm; therefore 1 metre = 99.9998 cm, alternately, 1 inch (international) = 0.999998 US inches.--Wendy.krieger (talk) 08:11, 28 February 2011 (UTC)
 * centi is 1/100, so 100 cm = 1 m, or 100 cl = 1 l (exactly). For the difference of exactly 2 ppm: It is in the numbers of the definitions - (Assuming 1 yard = 3 feet = 36 inches (exactly) both for "international" and "U.S. survey":) The international yard is defined as exactly 0.9144 metres. And The US survey foot is defined so that 1 metre is exactly 39.37 inches. So because of these 2 (exact) defintions the relation of the international foot to the U. S. survey foot is 1 ft (international) / 1 ft (U. S. survey) = ( 0.9144 m / 3 ) / ( 12 * 1 m / 39.37 ) = ( 3937 * 9144 / 3 ) / ( 12 * 1000000 ) = ( 3937 * 3048 / 12 ) / 1000000 = ( 3937 * 254 ) / 1000000 = 999998 / 1000000 = 1 - 2 ppm . q. e. d. --79.214.64.211 (talk) 23:08, 2 April 2013 (UTC)




 * 1 m ||= 39.37 US in||(by definition)
 * ∴||1 US in ||= 1/39.37 m
 * 1 int yd ||= 0.9144 m||(by definition)
 * ∴||1 int in ||= 0.0254 m||(∵ 1 yd = 3 ft & 1 ft = 12 in)
 * 1 int ft / 1 US ft ||= 12 int in / 12 US in
 * = 1 int in / 1 US in
 * = 0.0254 m / (1/39.37 m)
 * = 0.0254 × 39.37
 * = 0.999998
 * = 1 − 2 ppm
 * }
 * = 1 int in / 1 US in
 * = 0.0254 m / (1/39.37 m)
 * = 0.0254 × 39.37
 * = 0.999998
 * = 1 − 2 ppm
 * }
 * = 0.999998
 * = 1 − 2 ppm
 * }
 * }


 * The international foot is exactly 2 parts per million shorter than the US survey foot (as the article says) but the US survey foot is not exactly 2 parts per million longer than the international foot (for comparison 4 is 20% smaller than 5 but 5 is 25% bigger than 4). How about a proof in a footnote? J IM ptalk·cont 01:46, 3 April 2013 (UTC)

Not sure how NIST defines the international survey foot, nor do I particularly care, but here is the result of dividing 1/39.37 and represents the actual relation between U.S. Survey and International feet:

2.54000508001016002032004064008128016256032512065024130048260096520193040386080772161544323088646177292354584709169418338836677673355346710693421386842773685547371094742189484378968757937515875031750063500127000254000. ..

Notice the value is doubled until it repeats back to the same series after 210 digits. While 2 ppm, 4 ppt, 8 ppqu (1.000002000004000008000016. . . )etc. may diminish in importance rapidly, when measuring a particularly large distance with a high degree of resolution required, that differential could conceivably come into play. Practically 1.000002 is close enough, but that doesn't make it the definition of the survey unit compared with the international equivalent, 1/39.37 meter defines the survey inch. 1.000002 is an approximation.Surveyor792 (talk) 20:50, 27 September 2013 (UTC)

BW95 and Jimp have it right. 1.000002 is an approximation; the US survey foot is not exactly 2 ppm longer. The inverse ratio 0.999998 is exact; the International foot is exactly 2 ppm shorter than the US survey foot. The article should state "exactly 2 ppm shorter" and not "approximately". Surveyor792, your discussion confuses the issue. No further explanation is needed in the article.BillHart93 (talk) 01:23, 30 September 2013 (UTC)

Somebody still doesn't believe it. If you are using a calculator, 0.9144 meter/Int'l yard divided by (36/39.37) = 0.9999980000 That comes out 2 ppm less to calculator precision.

It's better to do the arithmetic manually to get an integer ratio:

[(12*0.0254) meters per Int'l ft ] divided by [(12/39.37) meters per US ft] = (39.37 * 0.0254) = (3937 * 254)/1,000,000 = 999,998 / 1,000,000

Note that if one is 2 ppm less, the other is not exactly 2 ppm more. The statement is correct that the International foot is exactly 2 ppm shorter than the US survey foot, but the reverse is not exact. BillHart93 (talk) 02:03, 2 October 2013 (UTC)

You just want to argue, don't you, to be the biggest cock in the roost. I'm agreeing with you, showing the exact number of ppm larger that the US survey foot is compared to international, and you say that this is "unnecessary." So let's just ignore the actual number and argue about the words"exacty" and "approximately." VERY productiveSurveyor792 (talk) 18:19, 20 October 2013 (UTC)


 * There are many ways to calculate the relative differences between two variables. Suppose that we call the variables A and B. Four ways of doing the estimation are:
 * Using A as the reference value: abs(A-B)/A
 * Using B as the reference value: abs (A-B)/B
 * Not favouring either A or B (method 1): abs(A-B)/2(A+B)
 * Not favouring either A or B (method 2): abs(A-B)/sqrt(A*B)
 * in the case in question we only get an relative difference of exactly 2 ppm if we use the US foot as the reference foot. Since there is no reason to favour either the US foot above the UK foot or vice-versa, we should use either the third or the fourth method described above. Neither gives exactly 2 ppm. I therefore stand by my changes. Martinvl (talk) 05:52, 2 October 2013 (UTC)


 * I think this argument is now more about English than about arithmetic. If I say B is lighter, shorter, cheaper or whatever than A by some extent then I do so with reference to A. So if I say that widget A cost 50 cents but widget B is 50% cheaper, I'm saying that widget B costs 25 cents. If I say the old widget weighed 1000 kg but the new one weighs exactly 2ppm less, I mean that the new one weighs exactly 999.998 kg. Likewise, if I read that the imperial foot is 2ppm shorter than the US survey foot, I learn that one imperial foot is 0.999998 US survey feet - which is true and even exact. NebY (talk) 07:49, 2 October 2013 (UTC)

Bill Hart: I took a great deal of time to manually input the exact decimal value of 1/39.37; there is no need to be so incredibly rude and dismissive. I'm the only one who has actually bothered to work out the actual numerical progression.

The numbers don't lie, like many of the "editors" on here who care more about stroking their own egos and being "right." How is showing that 1.000002000004000006. . . is the actual relationship not conclusive and definitive?

Come back on here and there are another twenty-five paragraphs of BS. Talk talk talk talk talk by a lot of people who are apparantly too lazy to scroll across the page and look at the numbers. This IS, after all, an article about numerical relationships and metrology. :-0 Surveyor792 (talk) 23:49, 12 October 2013 (UTC)

You have proven that the US survey foot is not exactly 2 ppm larger than the international foot (or yard/yard). That is not the ratio the article states is exact. The article deals with the reciprocal of your number. Multiply 0.0254 * 39.37 = 0.999998 exactly. That demonstrates the article's claim that the international foot is exactly 2 parts per million shorter than the US survey foot. BillHart93 (talk) 04:00, 13 October 2013 (UTC)

Merger proposal
I propose that a newly created article, Imperial and US customary measurement systems, be merged into this article. I think that as there is huge overlap and duplication between the two, that the content in the new article can easily be accommodated in the context of this article, and the new article is of a reasonable size that the merging of it here will not cause any problems as far as article size or undue weight is concerned. 212.183.140.15 (talk) 21:43, 11 July 2013 (UTC)


 * The section Talk:Imperial and US customary measurement systems clearly identifies the scope of that article which is to provide an overview for four articles, one of which is Comparison of the imperial and US customary measurement systems. Given that the article Imperial and US customary measurement systems is about 75 kbytes in length (as per WP:LENGTH articles should try not to go much above that value) it would not be appropriate to merge it into any article, let alone one that has a "Citations needed" banner across it and text that is riddled with "Convert errors". Martinvl (talk) 21:49, 12 July 2013 (UTC)


 * Comment: The new article is far superior in every respect, including coverage and sourcing. The 75kb guideline would not trouble me if a few additions were needed to make it complete. I would favor including in the new article any accurate information from the old article that is needed for complete coverage. It should be verifiable from sources already cited or new ones - because the old article is unsourced. I doubt the additions would be very large because of the noted considerable overlap of the two articles. Then the old article could be deleted. Perhaps the old article complete overlaps the new one and could be deleted without any carryover, but I have not checked the articles closely enough to endorse that. The suggested action technically would be a merger, I suppose, but it seems somewhat less than that because I think little would be carried over to the new article. In view of the overlap that is apparent even from a cursory look at the two articles, I think that only one article on this topic should exist. Donner60 (talk) 04:21, 13 July 2013 (UTC)


 * The "new" article is an overview for four existing articles, one of which is the "old" article. The other three articles are  English units (pre 1824), Imperial units (post 1824 - UK), United States customary units (post 1776 - US).  There is scope for the "old" article to be improved, starting with Thomas Jefferson's report in 1790. (Jefferson noted that there were already differences between the US and the UK, notably that not all UK units had been adopted in the US). Martinvl (talk) 04:36, 13 July 2013 (UTC)


 * For the record, I believe the proposer of this merger is a malicious move by the banned User:DeFacto. I have posted a request on the ANI noticeboard that this merge request be swiftly closed on those grounds. Until I have a positive second opinion that this is the case, I will treat this as a genuine, but misguided proposal. Martinvl (talk) 07:41, 13 July 2013 (UTC)


 * Support merge as per the proposer and Donner60 above. The new article has its merits, but as it covers, in detail and not just overview, most of the content of the original article, it should have been added to the original article and not in a newly created article. Also, with the new article, the intricacies of the history of the article development are lost. 78.46.43.39 (talk) 08:07, 13 July 2013 (UTC)


 * Now that User:Martinv has explained this in a manner that has finally sunk in with me, I see that this is supposed to be an overview article and the other article is to have additional details. The other article does not need to be merged or deleted but certainly needs to be improved and better sourced. Martinv explained this at the outset but I did not completely follow the explanation; I think it was because I lost the progression in switching back and forth between the talk pages. In any event, your article is quite good and the topic is quite important. Nothing should be lost in the process here and if retention of both articles is needed for the full explanation and details to be presented, I would favor that - with the eventual cleanup and sourcing as needed. It would by no means be anomalous for this to be the way to go, even though the topic of the two articles seems (superficially, as it now appears) to suggest one article would be enough. Donner60 (talk) 08:30, 13 July 2013 (UTC)

What convention?
The "US" in the title has been changed to "U.S." without any fuller explanation than "per convention". Imperial and U.S. customary measurement systems has been hit too. Who's convention? The MOS convention is to allow either. Jimp 09:07, 14 August 2013 (UTC)


 * I agree with Jimp. WP:MOS states quite clearly that "U.S." is the norm in United States and Canadian English, but that "US" is the norm in other varieties of English.  A cursory glance at the article identified British English as the variety of English used - consistent use of "US" rather than "U.S.", consistent use of "metre" rather than "meter" and the absence of a comma between the words "quarts" and "or" in the phrase "the bushel into 4 pecks, 8 gallons, 32 quarts or 64 pints".  In view of this, please change the title back to it UK English form. BTW, I am about to mark this article as using UK English. Martinvl (talk) 09:42, 14 August 2013 (UTC)