Scientists think our recent ancestors may have run hotter -- but why?

150 years ago, average human body temperatures appear to have been higher than now. Scientists are still exploring exactly why.

In the 1860s, the German physician Carl Wunderlich, was faced with a burning question. Was fever its own disease, or a symptom of an illness underneath?

At a time when thermometers were not routinely used, Wunderlich embarked on an ambitious mission. Over 15 years, he gathered the armpit temperatures of 25,000 patients, of all ages and both sexes, resulting in over a million readings.

From this vast dataset Wunderlich calculated an average, 37C (98.6F), for healthy adults. This became the benchmark for tracking normal human body temperature. If that rose above 38C (100.4F), it signalled fever, he wrote.

Even though others had already arrived at the same conclusion, Wunderlich's dataset was so large that his figure has since enjoyed an "exalted status" as a universal diagnostic tool, says Philip Mackowiak, a medical historian and infectious disease expert at the University of Maryland.

But more recent research suggests that the average normal body temperature has also changed over the last century and a half. Was the original baseline wrong? Or is a fast-changing world changing our biology?

Wunderlich "established a lot of what we know to be true" about body temperature variation says Julie Parsonnet, an infectious disease clinician at Stanford University.

From his dataset, which measured mostly working-class patients at the Leipzig hospital where he was medical director, Wunderlich discovered that normal temperature varies. He noticed that women have higher temperatures than men, temperature declines with age, and also increases with the time of day. 

Yet the average individual today has also now been shown to be cooler than their counterpart 150 years ago, with American men running about 0.59C lower than in the early 19th-Century.

A decade ago, Parsonnet's team analysed three large datasets involving multiple temperature readings from nearly 190,000 people in the United States, taken between 1862 and 2017. "The new normal [for adults] seems closer to 97.5F (36.39C)," she says.

Numerous studies, both before and since, echo Parsonnet's findings.

Nor is the downward trend limited to people living in developed countries. The average body temperature of Tsimanes, a semi-nomadic indigenous community living in the Amazon rainforest, used to hover around 37C in the early 2000s, says biological anthropologist Adrian Jaeggi at the University of Zurich in Switzerland. But in 2018, following two decades of rapid socioeconomic change, their average fell to 36.5C (97.7F).

A handful of explanations could explain the lower averages we see today. Parsonnet, for instance, believes, they are the result of healthier overall populations. In Wunderlich’s day, life expectancy was 38 years. Today, it's nearly double that.

"People didn’t live the healthy lives that we live now," says Parsonnet. "They [often] had lots of things wrong with them – tuberculosis, terrible dental care, dysentery, rheumatic fever, and osteomyelitis."

"When your immune system is trying to defend you against all these things that are trying to kill you, it consumes calories. And calories are another word for heat," Parsonnet says.

Likewise, with the Tsimanes, vaccines and improved sanitation have improved health, says Jaeggi, Toilets and shoe-wearing are also more common now than 20 years ago, he notes, reducing the number of hookworm infections acquired by walking barefoot on contaminated soil.

Indoor air temperatures are also now easier to control. Air conditioning is present in at least 88% of US homes, while a similar share of UK houses have central heating.

"Most of our lives are spent in a perfect climate, in what’s called a thermal-neutral zone where we don't have to work to maintain our temperature," says Parsonnet. "And if you don’t have to work, you don't produce energy, and you don’t produce heat."

Mackowiak, however, thinks the fall in average body temperature might be due to something else entirely.

When he had the chance to examine one of Wunderlich's glass thermometers in person, Mackowiak told me he noticed that it was calibrated roughly 2C higher than modern thermometers. This means that any readings made via these instruments would be 2C warmer than they should have been.

Back then, measuring temperatures were a cumbersome affair, says Mackowiak. Mercury-in-glass thermometers, roughly a foot long, had to be first calibrated with fixed points such as boiling water and melting ice. Patients would then have to lie still for 15 to 20 minutes because "it took a while for the temperature to register in those thermometers," he said.

In other words, Wunderlich's figure might have been flawed to begin with. He also took armpit temperatures, rather than the oral readings that are more common today – making his measurements tricky to compare to modern ones.

Parsonett, however, doesn't think faulty calibration is to blame for Wunderlich's overestimate. When she and her team analysed body temperature data collected from Union Army veterans between 1862 and 1930, they found that those born earlier did indeed have higher average temperatures than those born later.

"Because we compared people born in different years who were measured at the same time, and people born in the same year measured at different times, you can separate out the instrument [effect] from the age and birth year," she explains.

Parsonett's team was further convinced when they looked at two other, more modern, datasets – and saw a similar drop. These comprised temperatures taken between 1971 and 2017, "using thermometers that would be expected to be similarly calibrated".

Does all this mean it's time to rethink the special significance accorded to 37C (98.6F)?

Some already have, including two authoritative sources widely used by physicians. This includes the book Harrison's Principles of Internal Medicine and the UptoDate online database.

Yet while having a benchmark figure for your own body temperature can be a helpful guide, most experts caution against putting too much emphasis on a single value.

More like this:

• The radical idea that temperature sets when we sleep

• The world is getting hotter - this is what it is doing to our brains

• Should you always treat a fever?

"It's important not to take these kinds of thresholds too seriously as they don’t represent sharp boundaries – no one would argue that 36.99C is somehow radically different from 37.01C," says Jaeggi. 

Still, when temperatures drop too low or spike too high – typically below 35C (96F) or above 40C (104F) for adults – it is an indicator that something more serious could be going on in your body. 

* All content within this article is provided for general information only, and should not be treated as a substitute for the medical advice of health care professionals.

--

For trusted insights on health and wellbeing, sign up to the Health Fix newsletter by senior health correspondent Melissa Hogenboom who also writes the Live Well For Longer and Six Steps to Calm courses.

For more science, technology, environment and health stories from the BBC, follow us on Facebook and Instagram.