At any rate, this is the bit I was polishing. It finds me in the role of weatherman taking his turn on the ten o'clock news. "Well, there's good news and bad news," I would say. "The good news is it looks like this dreadful cold snap is finally coming to an end. Tomorrow the thermometer is expected to climb back above the freezing point for the first time in weeks, and on Saturday we should reach the balmy 40s. And on Sunday—break out your golf clubs! They say it's going to be sunny in the 60s.
"The bad news is that the earth is falling into the sun. The Weather Bureau says that by this time next week we should be looking at a high of about 14,000 degrees."
The anchorman breaks in. "Is that Fahrenheit or Celsius?"
"Darned good question. I'll see if I can find out by the end of the broadcast. So Biff, did the Blackhawks keep their streak alive?"
Then I got home and picked up the New York Times and ran into some intriguing synchronicity. I came across the awful news that the International Herald Tribune is going to be renamed later this year the International New York Times. Having squeezed the last ounce of romance and tradition out of its appendage, the Times could afford to reminisce. Said the account by Sam Roberts:
The quirky paper based in Paris reflected James Gordon Bennett Jr.'s eccentricities (printing for 6,718 consecutive issues a bogus letter signed "Old Philadelphia Lady" that explained how to convert Celsius into Fahrenheit and vice versa).
I had never heard of this letter, and I immediately wondered what her explanation was.
Thanks to the Internet, the letter isn't hard to find. Roberts mischaracterized it. She didn't explain. She was merely asking. Her letter said in its entirety:
To the Editor of the Herald:
I am anxious to find out the way to figure the temperature from Centigrade to Fahrenheit and vice-versa. In other words, I want to know whenever I see the temperature designated on a Centigrade thermometer, how to find out what it would be on Fahrenheit's thermometer.
OLD PHILADELPHIA LADY.
Paris, Dec. 24, 1899
This letter was discussed by the Herald Tribune decades later in an anniversary issue. The paper explained that Bennett's purpose had been to persuade the Anglo-Saxon world to adopt the Centigrade [Celsius] thermometer, a campaign that proved to be a "lamentable failure" and was dropped after Bennett died in 1918.
The article said that over the years hundreds of replies to the Old Philadelphia Lady's query rolled in, some helpful, others not. Then the Herald Tribune itself weighed in. "A handy way to make the conversion," the article went on, "is to multiply the Centigrade temperature by nine and divide by five, adding 32 to the quotient. If the Centigrade reading is below 0, then all that's needed is to multiply by nine and divide by five."
This was an attempt to offer readers the same conversion method I learned in school as a boy—but the attempt went horribly wrong.
What we learn in school is that water boils at 212°F and 100°C. It freezes at 32°F and 0°C. And -40°F = -40°C. Use the above formula, and 100°C converts to 212°F and 0°C converts to 32°F. But -40°C converts to -72°F. It doesn't matter whether the Centigrade reading is above or below 0. After the multiplying and dividing, the 32 must be added.
And what about converting F to C? Here's where it gets really tricky. First subtract 32 from the Fahrenheit reading, then multiply by five and divide by nine.
Remembering whether to add or subtract the 32, and whether to do that first or last or at all, can confound the finest minds. Such as the author of the Herald-Tribune's anniversary article, who fell out of his canoe in midstream.
Here are the formulas:
Fahrenheit [°F] = [°C] × 9 ⁄ 5 + 32 [°C] = ([°F] − 32) × 5 ⁄ 9
As I say, I learned them in school. That was several years ago, but I just copied these formulas from the Wikipedia entry on Celsius. Apparently, they're the ones people still use.
They should be obsolete. In college, a professor teaching something unrelated confided to his class—with an air that suggested we all pray the room wasn't bugged—that there's a much simpler method. It's based on the fact that -40 degrees is the same in both measurements. So to whichever reading you want to convert, add 40. Then multiply: by 9/5 if you're converting C to F, by 5/9 if you're converting F to C. Then subtract 40. 40 is an easier number to work with than 32, and the order of the steps you take is always the same.
Does anyone but me know this? My quick, dirty Internet search suggests no. Here's a TV weatherman in Seattle weighing in:
Those who travel back and forth to Canada, or other countries on the metric system, will find temperatures in Celsius.
The official way to convert C to F is to multiply C by 1.8, and then add 32. But if you'd like a quick way to get a rough guesstimate and give your brain a break from multiplying by 1.8, you can instead take your C reading, double it, and then add 30. It'll get you within a few degrees. (20 degrees C * 1.8 = 36, add 32, = 68 degrees. Quick way: 20*2 = 40, plus 30 = 70 degrees.)
To go from F to C, you take F, subtract 32, then divide by 1.8. Or the easy way: Subtract 30, then take half.
The official way! Is an international treaty involved?
Knowing sub rosa the better way to convert temperatures has long made me feel a little special. Finally it's cracked the comedy routine. It's part of the act.