One thing thats always bother me about farenheit is why is freezing 32? If you’re going to start your own scale wouldn’t you want the references to be easy to deal with? I mean 0 and 180 work much better than 32 and 212.
Just read Why do we have so many temperature scales?, I take it?
Well, you’re in luck, the answer to your question can be found here.
First, congrats to hibernicus on a great Special Report!
Second, welcome to the Straight Dope Message Boards, genaldar, we’re glad to have you with us. It’s helpful to others if you provide a link to the Column or Special Report that you’re commenting on, saves searching time and keeps us on the same page. No biggie, Kat has helped you out, you’ll know for next time.
Third, you might want to explore around the site, and especially the section called Archives, where you will find many questions answered. Cecil has been in the business for thirty years plus, and we’ve had additional Staff Reports for five or six years. So lots of questions have already been answered, plus hours (days?) of wonderful reading time.
So, welcome, we’re glad you’re here!
Could somebody please point out to me the (blindingly obvious, I’m sure) reason that Celsius and others felt that an inverse scale (with 0 being boiling and 100 freezing) was a good idea? There must have been a practical reason for this to have made sense to people, right?
Also, oftentimes people refer to Celsius degrees as “Centigrade”- when did this become common? Is this a Britishism? I fully understand the term, and why it applies to the Celsius scale, but when did people transition to this terminology, and where?
Thanks,
-j
The article implies the scale was originally called “Centigrade”, and was only officially named Celsius in 1948. Therefore, any usage of the term “Centigrade” is a holdover from the old days.
One nitpick though, the use of the units’ symbology isn’t all consistent in the article. While the Kelvin scale doesn’t use the degree symbol, others do. That’s because the kelvin is a unit of measurement, while the others are degrees of a scale of measurement. And in SI, “kelvin” is not capitalized when spelled out.
So water boils at:
373 K (373 kelvin)
100 °C (100 degrees Celsius)
212 °F (212 degrees Fahrenheit)
672 °R (672 degrees Rankine)
I can’t say for certain how the other scales work. But that little bit of factual inaccuracy caught my eye.
Fighting ignorance and all…
Hibernicism,
I seem to me that humanity’s quest for absolute temperature measurement has left us with a scale for everyday temperature measurement that is compressed with me wondering if I have a slight fever because the thermometer changed seven tenths of a degree Celsius or do I need to run to the doctor as the thermometer has now changed four degrees Celsius. If a real-world temperature scale necessarily be considerably less elegant as you stated why not use a scale that can be quantified in easier terms for everyday life? Sure let’s find out exactly when an atom stops all sub atomic movement and use the proper scale for that measurement, but lets cut us everyday humans some slack and use a scale that is easily read.
Because at the time the concepts of conservation of energy (and of heat as energy) and of absolute zero were both somewhere between vague and nonexistent, and one set of arbitrary marks was as good as another. There’s no law that says that house numbers have to go from north to south or south to north, east to west or west to east, and school grades can run A, B, C, D, F, or (at my junior high school) A, B, C, D, E, or (in my elementary school) E, S, N, U.
We understand now that heat is energy (he said, carefully avoiding a segue into Flanders and Swann) and can be quantified as such; and that cold is merely the absence of heat; but that wasn’t the hare that Celsius was hunting. (In fact, the very existence of the International Temperature Scale today shows that the strictly energy-quantitative Kelvin scale isn’t altogether a practical one.)
I know some discount this as hoo-hah, but if you know the details it is not.
The freezing point of water was AN APPROXIMATION of when harbors and bays would freeze up (that is not the same as the freezing point of salt water…).
The freezing point of such an event is quite difficult to determine; it will be dictated by salinity and currents. But I have found that when it’s zero Fahrenheit for more than a few days, the harbors here in New York are usually frozen, or get frozen fast. Such an indicator is somewhat useless if you are inland and only live near bodies of fresh water. But it’s handy for coastal dwellers.
I am generally an advocate of the metric system, but I think on the thermometer it’s not really quite practical. In the west, we generally find it convenient to express things as percentiles, and most temperatures experienced by us (unless you’re in Vladivostock or Death Valley) are generally in Farhenheit’s 0-100 range. I find it more cumbersome to always add syllables for Celsius (e.g., “21 point two” as opposed to something simple like “87”).
But (nitpick again)… thermometers don’t measure heat. They measure temperature. An iceberg has a lot more heat than a burning match.
I know you don’t exactly say that temperature = heat, but your post comes very close to implying that it’s heat we’re worried about. The nitpick doesn’t make a difference in the spirit of what you were saying, but I thought it should be clarified.
Sorry; no edit capability as a visitor!
My opening statement “The freezing point of water…” should be “the 0 degree mark of Fahrenheit…”
Actually, you can make water boil at any temperature you want by simply varying the air pressure.
In the article, excepting the Question, the word “Kelvin” is used in the following ways:
“…are the Celsius and Fahrenheit scales, and (mainly for scientists and engineers) the Kelvin and Rankine scales.”
“…the Rankine scale, based on the Fahrenheit degree, and the Kelvin scale, based on the centigrade degree. Note that the SI unit of temperature is called the “kelvin” (not “degree kelvin”). The freezing point of water is 273.15K and 491.67Ra.”
“…the Kelvin (and hence, Celsius) temperature scale was redefined in 1954…”
In which case is “Kelvin” or “kelvin” mis-used? They all appear correct to me. Every place he uses “Kelvin” he’s capitalizing it in reference to the name of the scale, not as a unit. There are some examples on this link below.
Touché! You got me there. But my point still stands.
And, just to counter your pickiness 
… You can’t make water boil at any temperature. At any temperature below the triple point, water can only transition between solid and liquid. Rapid sublimation, yes. Boiling, no.
*And of course, I know there isn’t just one triple point for water. Since there are several exotic and metastable phases of water (mostly odd crystalline structures for ice), there are multiple triple points. I’m referring to the common triple point between hexagonal-crystal ice, liquid, and vapor. *
I wasn’t saying it was…I was just adding that point, almost as an aside, while I was talking about the temperature scales. Sort of emphasizing a point not given explicitly in the column–a point that may have been lost on the casual reader. I was in no way saying that the author misused the unit/name.
I was, however, making a point about the use of the degree symbol and inclusion of the word “degree” in the unit name. The article doesn’t follow rigorous conventions for unit labeling. Is it a big deal? No. Do I sometimes omit the degree symbol in my engineering work? Yes.
Your point is valid, aerodave, and in fact there were degree symbols in the original text of my article, but somewhere in the communication process or the transition to HTML they dropped out (or perhaps they were eaten by hungry hamsters). On the plus side, though, I see that Ole Rømer kept his Danish ø.
smoke, there is no compelling reason why Celsius created an “inverted” scale. It was an entirely arbitrary decision.
Celsius explains in his paper of 1742 the process by which his scale is set up. He chose as his upper fixed point the temperature of boiling water (with large bubbles over the entire surface of the water) when the barometer stands at 25.25 Swedish inches (=751.16 mm). He marked this as zero, and the temperature of melting snow as 100.
The question of who first suggested that the scale should run in the other direction is a subject of considerable controversy, although it certainly happened by 1745 or 1746. One possible candidate is a Stockholm instrument maker named Daniel Ekström. Others include Linnaeus, Märten Strömer and Jean-Pierre Christin (whose claim is favoured by French writers). At any rate, the original Celsius scale was published in 1742, and was no longer used from 1750, having been replaced by Strömer’s version (with 0 = cold, 100 = hot). Anders Celsius died in 1744.
In the case of the Delisle scale, the reason arose from the fact that he used a single fixed point: the temperature of boiling water. This he called zero, and graduated the tube downwards in terms of the contraction of the mercury, expressed as hundred-thousandths of the original volume. He later changed it to ten-thousandths, to make the numbers easier to deal with, and as it happened the temperature of freezing water turns out to be around 149.5 on this scale.
I will let Fahrenheit himself answer that question:
Fahrenheit, Phil Trans, 33 (1724) pp 78-84, quoted in Middleton (1966) A History of the Thermometer and its Use in Meteorology
However, there are reasons to distrust the truth of Fahrenheit’s description. For one thing, he mentions two different salts, which casts severe doubt on the value of this fixed point.
A far more likely explanation is that he simply used the zero point from his own version of Rømer’s scale, and that the original Fahrenheit scale actually had two fixed points, and not three as he implied in his paper.
We should also note that the boiling point of water was not originally a fixed point, and was measured by extrapolating the scale upward in proportionate length from 96 (which is a highly unreliable endeavour). However it quickly became common to make “Fahrenheit” thermometers using boiling water as a fixed point with the value 212.
JonG, it should be said that any scale which has survived in use for over 250 years (like the Celsius and Fahrenheit scales) have proved their real-world usefulness. Certainly there is no immediate danger of Kelvin replacing Celsius or Fahrenheit on your medical thermometer - it would be ridiculously unwieldy.
Having said that, there is little to choose between Celsius and Fahrenheit for the purpose of determining whether you have a fever. Which one you use depends primarily on where you grew up, rather than on any inherent advantage of one scale relative to the other.
Logicom, I have no interest in getting involved in a discussion of which sclae is better, but I don’t accept either of the points you make for the superior utility of Fahrenheit. The fact that 0 to 100 on the Fahrenheit scale corresponds well to meteorological temperatures says more about where you live than about the merits of the scale. And the idea that Celsius is more awkward because you have to add syllables (“21 point two” as opposed to something simple like “87”), is wrong - when talking about the weather, people just round to the nearest degree Celsius instead of the nearest degree Fahrenheit.
Hibernicus,
First, great staff report. Very clear and informative. My only (minor) nitpick turns out to be a case of hungry hamsters. 
I have a question similar to Smoke’s
In the article you report that the name was officially changed to “Celsius” in 1948. Perhaps this is just an “Americanism”, but from primary school (ca. 1965) until the late '80’s, I always heard the terms centigrade and Celsius used interchangeably. A number of my science teachers insisted on centigrade as the preferred term, because it was descriptive, denoting a 100 point scale. However, sometime during Engineering school, Celsius became the proper term because it reflected the person involved with it’s discovery, e.g., Newton, Tesla, Guass, Joule, etc.
What is the straight dope on the name? Did it swap back and forth, like Cape Canaveral/Kennedy/Canaveral, or was I just educated in a cultural backwater, i.e., Texas?
Not that it matters, since this excellent Staff Report far more than stands on its own merits, but as one of hibernicus reviewers I can confirm the degree symbols were only lost during the Report-HTML transition.
Thanks for answering my questions, and that really was a well-written areport, hibernicus. So since I’m not the only one wondering (thanks, Rhubarb), is the centigrade/Celsius thing regional, a matter of the times, or what? Is there any strong feeling in the scientific community as to which should be used? I would think Celsius is “more correct”, because there have been other centigrade temperature scales, but (and this may be from listening to BBC news- or not) “centigrade” just strikes me as being British, only without extraneous syllables.
Don’t get me wrong…I thought it was a great report. As I said in my first post, it was being exceedingly nit-picky. Kudos to hibernicus on a thorough and well-written article.
Just one more question, out of curiosity: How do members who are not on the SDSAB, such as hibernicus, get questions to answer?
I imagine that there are several areas of expertise not fully accounted for in the SDSAB pantheon. Is it common for these questions to be referred to members with known credentials in the appropriate subject area?