{♫Intro♫}.

If you're coughing, sneezing, or starting to feel under the weather, you might blame a virus, or possibly a bacterium. Which is not something humans have known to do for very long.

Around 400 B. C. E., doctors might have blamed an imbalance of the four vital humors for your illness.

Around the 1700s, they might have pointed to an invisible, disease-carrying fog instead. Today, we know pathogens -- viruses, bacteria, and certain other microbes -- are responsible for many diseases. But linking specific diseases to the microbes that cause them has been surprisingly tricky.

In 1882, a scientist named Robert Koch demonstrated that the bacterium Mycobacterium tuberculosis causes tuberculosis. And in 1890, he also published a framework for future scientists to make similar discoveries. He created a checklist for researchers to reference any time they're trying to link a pathogen to a disease.

The steps are as follows. First, researchers had to be able to find the pathogen in sick organisms, but not healthy ones. Second, it could be grown in pure culture -- which means that a sample of the microbe could be taken from a sick organism, and then the microbe could grow independently in a 19th century version of the petri dish.

Third, if they exposed a healthy organism to the stuff that they grew in step two, that organism would get sick with the same disease. Finally, though this step is sometimes considered optional, the same microbe that was isolated in step one must be found again in the organism made sick in step three. These steps are now known as Koch's Postulates.

The idea is that if the microbe meets all of the postulates, then you know it's the cause of the disease. Unfortunately, his postulates had a few problems. Take postulate one.

Tuberculosis can actually be found in healthy individuals -- that's called latent tuberculosis -- so it doesn't meet Koch's first postulate. This situation just didn't show up in his experiments, which were done in guinea pigs. Postulate three isn't perfect either.

Assuming that any healthy organism exposed to a pathogen will get sick ignores differences in immune systems. A healthy organism might be able to fight off the infection or might already be immune to the disease. But it was the second postulate that caused the most confusion.

Something grown in “pure culture” has to be the only living thing in the dish, and many pathogens just can't grow independently like that. Viruses, for example, reproduce by hijacking molecular machinery in the cells of the organism they're infecting. Meaning you can't grow them in a dish by themselves.

But bacteria often grow in a dish just fine. Because postulate two required the thing to grow in culture, researchers at the turn of the twentieth century would almost exclusively blame bacteria for the diseases they were studying, which resulted in some false accusations. Malaria, which is actually caused by blood-infecting parasites, was blamed on a bacterium from.

Italian marshes in the 1880s, which they named Bacillus malariae. Canine distemper, a sometimes deadly disease in dogs that causes symptoms like fever and vomiting, was linked to a series of different bacteria before it was finally proven to be a virus in the 1920s. And the familiar virus influenza, or the flu, was misidentified as a bacterium in eighteen ninety-two, by a colleague of Koch's.

The bacterium came to be known as Haemophilus influenzae. To study the flu, researchers needed samples of spit and snot from people with obvious symptoms. But one thing that made influenza hard to study was that, even though the flu usually reaches a peak in winter, the only time that scientists could reliably find large numbers of flu-ridden folks at the same time was during a pandemic.

And those could be decades apart. So the first chance scientists had to check the results from 1892 was during the next influenza pandemic… in 1918. Researchers were unable to replicate those initial results.

But it wasn't clear at the time if it was because of poorly controlled studies in the chaos of one of the worst pandemics in recent history and the end of World War I, or if they were just... wrong. A vaccine was developed in New York based on Haemophilus, just in case. There was at least one study around that time that managed to find evidence of the right answer: influenza is a virus.

It took until 1933 and another influenza pandemic for scientists to prove without a doubt that the flu is caused by a virus, thanks to the introduction of ferrets as a model organism. Ferrets were the only small mammals they could find that actually get the flu and show symptoms similar to ours. So it seems like Koch's Postulates, especially the second, really hindered research into any disease that didn't have a bacterium behind it.

Does that mean they're useless? Not at all. Since the 1880s, scientists have tweaked Koch's postulates over time to match modern understandings of pathogens.

Today, the focus isn't just on microbes, but on their genes. Using genetic sequencing, scientists can gather information about all of the nucleic acids in a sample, whether DNA or RNA, and then use a modified version of Koch's postulates to figure out which genes are most associated with disease symptoms. For example, in 1996 scientists at Stanford came up with a new set of postulates with seven gene-centric points.

By using gene sequencing, scientists can find pathogens that haven't been isolated and identified before. And there's no need to culture them. Koch's postulates provided a solid foundation for researchers to begin linking diseases to their sources.

Sure, there were a few mistakes, but they provided a rigorous, testable basis for understanding disease. Even if we had to come along and make some changes later. And even if some ferrets had to get the sniffles.

Thanks for watching this episode of SciShow, and thanks to our supporters on Patreon for making what we do possible. If you want to join them and help us, while also getting some cool benefits, check out patreon.com/scishow. {♫Outro♫}.