Back around the turn from the 17th into the 18th century there would have been no real science supply stores.  Most equipment would have been made by local artisans, glass blowers, blacksmiths and the like.  The problem facing researchers who wanted to explore patterns of changes in temperature and their effects was to set up a scale that everyone had access to.  They had to somehow normalize the data…find a base that everyone could set for themselves and then find another readily reproducible point.  Clearly this had not been 32 and 212; such numbers are too arbitrary.  What then?

The first thing to recognize is that 0° Fahrenheit is pretty cold, well below freezing.  Is there anything that happens on a pretty regular basis that is this cold?  There is a clue in the simple act of spreading salt to melt icy patches on your steps or sidewalk.  I’ve always been puzzled by this.  Spreading sand made sense.  It simply provided a grip for your step.  But salt isn’t going to lower the temperature.  How could it melt the ice?  It turns out, and you can try this at home, that salty water needs a colder temperature to freeze than clean water.  When we throw salt onto ice, we’re counting on the smallest crystals of salt mixing with odd bits of liquid water and so the surrounding ice finds itself not cold enough to stay frozen.

So, we may ask, at what temperature does salt water freeze?  It turns out that it depends on how salty the water is.  But if we use a salinity like that of ocean water, a sort of natural saltiness, then it freezes at 0°.  Pretty sweet.

With this as a baseline, Fahrenheit then looked around for a second ‘natural’ temperature.  What temperature would everyone have ready access to?  A clue is a temperature pretty close to 100°, the 98.6° of body temperature.  Fahrenheit had actually called this 100 and not 98.6.  His temperature scale had originally been a centigrade scale!  When scientists decided that freezing clean water was more reliable than the vagaries of ocean salinity and that the boiling point of clean water would be another equally secure and reproducible temperature…which happened to fall at 212°…that shifted things a bit and the ‘normal’ temperature for most of us most of the time became 98.6.

Reason restored.