This question is closely related to, but more specific, than the question/comment that inspired it: Could a modern times time traveler produce an antibiotic in medieval times?.

Imagine a person is sent back in time to roughly middle-ages or slightly after, without warning to prepare for it. He has a high school and college education, but not in biology. Perhaps he a mild interest in science/biology; but ultimately is mostly limited to what is taught, and he retained, from college level intro to biology class.

He wants to create penicillin, for obvious reasons. He knows the general story of how it was discovered, specifically by accident on a loaf of bread and it was noticed that the mold killed bacteria near it (or at least that's how he remembers the story, it doesn't mater how accurate that memory is since that is what he is working off of when making plans). He has decided to use the below approach to try to discover and recreate it. Assume he has the funding of a king that he impressed with modern technology, his budget is not limitless since the king isn't convinced that he can recreate anything, but it's sufficiently high for him to work with assistance and all reasonable expense covered.

To try to create penicillin he first gets many types of bread, and other wheat and food products, and intentionally lets them get moldy. He uses multiple different locations to grow the mold, mostly in dungeons but some in other places that aren't dank and moldy just in case penicillin grows better in other locations. He also gets the local blacksmith to construct very basic petri dishes and starts to grow many different bacterial cultures for testing.

Once bacteria and fungus samples exist he will place part of his bacteria samples in a separate dish with some fungus. He will watch to see if the bacteria appears to die out. He doesn't yet have a proper microscope since he hasn't figured out what to ask for to build one yet (When could a microscope first be made?). However, if he cultured bacteria well there will be enough built up in his petri dish to be able to see their presence with the naked eye, and thus tell when they appear to be killed.

He also gets a number of poor saps to play guanine pig if needed. Perhaps he gets partial pardons for criminals that would have be hung otherwise if they agree to play test dummy. He isn't unethical enough to do anything he considers to have a high chance of killing anyone, but is willing to use them as test subjects for what he considers lower risk human trials, especially considering how much good the drug will do if successful.

If he gets a fungus that appears to be antibacterial he will have some of this test subjects try it to see rather it makes them ill. He tries both oral consumption and exposing it to an open wound. He would start in very tiny quantities and only work his way up to higher levels if his subjects appear unharmed, generally he is trying to treat the subjects well and make their working for him be a clear and significant benefit over the alternative.

He will use separate controls later to test larger quantities of the fungus to ensure his original subjects didn't build up unnaturally high immunities to it from constant exposure to lower quantities. He is also aware that test subjects may develop germs with immunity to his fungus and so will use different subjects for testing the effectiveness of his fungus in fighting disease.

Assuming positive results he moves on to larger controlled trials with peasants, in the same way that we would do control trials with a new antibiotic today, other then the lack of infrastructure and communication systems making it harder to do quite as large a scale as he would prefer.

He generally takes his times, uses controls (as much as he can given the small number of samples/subjects he is working with), and triple checks any apparent result before moving on. However, while he has a decent understanding of the scientific method and intelligent he is not a scientist, and may still forget to control for certain variables or otherwise make mistakes that come from doing a job he was never taught how to do properly.

My main question is, would this work? What are the odds of his discovering any antibiotic fungus, that is relatively safe for humans, using this approach? He doesn't know what he needs to do to recreate penicillin, and I doubt he would find the actual penicillin fungus, but surely there are other antibiotic fungus he could discover which would suffice?

Assuming the main question is answered, and that this approach can work, here are some follow up questions that would be nice to have feedback on. The remaining are not mandatory to answer this question, I can post separate questions for them if need be, but any feedback that one wishes to add addressing them is welcomed.

  1. what are the odds of negative results of this experiment, and what sort of result could happen? How badly will his test subjects suffer from eating their moldy bread?

  2. Even if he finds a good fungus how easy would it be to then culture, grow, deliver, and teach others how to use it such that it would prove effective in fighting infection? Would it have to be reserved for rich high-paying nobles because they can't culture enough for regular use? He will try to prevent overuse creating penicillin immune bacteria, but the implications of such immunities are off-topic enough to likely warrant their own question later.

  3. how long would it be expected for this to take? Obviously this is more then a month's effort, but are we talking a year, a decade, or a lifetime?

  • 1
    $\begingroup$ Also, the Sliders did a version of this. $\endgroup$ Commented Aug 26, 2015 at 22:00
  • $\begingroup$ @DeveloperWeeks yes I remember it. However, in that case the presumed an expert who knew all about the process, to an unrealistic level in all honesty, considering he was not a medical professor, simply recreating steps he memorized. I'm more interested in how one does this if they aren't presumed to have perfect memory of the entire process. He knows the endgoal, but has large gaps in his knowledge of the actual steps; like a normal informed & intelligent laymen would have about the majority of our current knowledge/science. $\endgroup$
    – dsollen
    Commented Aug 27, 2015 at 14:38
  • 3
    $\begingroup$ In 1928 when he accidentally discovered penicillin, Alexander Fleming had a university degree in biology and 20 additional years experience using that degree where he was already well-known and regarded as a brilliant researcher. Even with all that, he spent 12 YEARS failing to come up with a way to make it practical to produce it and use it for treating humans. Some equally brilliant Oxford chemists discovered how to push the research forward in 1940 and worked with Fleming to make a usable antibiotic. Unfortunately Monty Wild is right; your intrepid layman traveler has no hope at all. $\endgroup$ Commented Feb 8, 2016 at 21:41
  • 1
    $\begingroup$ Antibiotics? He... if he want to improve the health of medieval people he could start with sanitation. It's a waaaaaay easier concept and it could still take a lifetime to convince others to adopt it. $\endgroup$
    – xDaizu
    Commented Jul 4, 2017 at 10:04
  • 1
    $\begingroup$ While penicillin is indeed a doomed effort, why not start with something simple? Mixing wet bread and spider web for treating wounds was popular in ancient times and rediscovered in 1600s. You can reasonably introduce it in medieval times to great effect. That and washing hands will make your guy pretty much into miracle worker. $\endgroup$ Commented Jan 29, 2018 at 8:10

3 Answers 3


TL:DR: This is going to be very difficult, if not effectively impossible. At any rate, if the individual in question does not have at least a solid grounding in biology, and preferably at least a year of modern university-level microbiology, they are most likely to spend a lifetime re-creating the tools needed to even begin to search effectively for any effective antibiotic, let alone penicillin.

Here are the problems:

Growth Media

Bacterial cultures don't generally grow on just anything. To get most strains or species to grow, you need a growth medium with basic nutrients added. I did Microbiology 101, which was enough to tell me what nutrient agar is, and how to prepare it (follow the instructions on the tin), but at that level it was only glossed over how to actually make it from scratch. Then, there are other types of agar, each designed to sustain fastidious bacteria with particular requirements or to assist in identification. You may well not even be able to grow some types of infectious bacteria without specific types of agar, let alone test them for antibiotic resistance.

Then you have the problem of the moulds you are testing. They, too, will require suitable growth media, and for the most part nutrient agar will be suitable, though some valuable strains may be more fastidious, and may require special media.

Equipment and Techniques

The equipment used in microbiology labs is fairly basic for our society, but is not nearly so basic for a medieval society. Pretty much everything in a lab takes place around a Bunsen, Teclu or Meker burner, which is commonly used to sterilise things such as inoculation loops, which have wide application. Inoculation loops require wire that will not corrode, even at high temperatures. These burners wouldn't even have fuel readily available in the medieval period, and even a wealthy noble might baulk at being told that (very pricey) fine platinum wire is required.

It takes a whole semester at Uni to train a student how to perform the gamut of basic microbiological techniques such as streaking (to isolate genetically pure strains of bacteria) without making mistakes that result in contamination.

As one example, to streak an agar plate to isolate individual bacterial species from a basic liquid culture of mixed bacterial species, the microbiologist must:

  1. Set up their burner so that it is between themselves and their equipment so that the flame sterilises their exhaled air - otherwise the plates are likely to be contaminated by the microbiologist's breath-borne bacteria.

  2. Set the burner to produce a hot blue flame. Yellow flames deposit soot on the inoculation loop, which contaminates the plates, and is not considered hot enough to kill some heat-resistant bacteria.

  3. Sterilise the inoculation loop by putting it just outside the blue part of the burner flame. It is a common beginner's mistake to put the loop inside the blue cone of a Bunsen flame, which is quite cold. The wire loop must be heated cherry red ("They're not dead until it's red!") or better still, orange, to kill any contaminating bacteria. Cool the loop by quenching it in an uninoculated portion of the plate. Failure to do this results in the heat killing the bacteria that it is intended to move.

  4. Dip the inoculation loop in the liquid culture and rub it across about a third of an agar plate.

  5. Sterilise and quench the loop. Failing to do this results in an undifferentiated bacterial lawn.

  6. Streak several times from the initial inoculation to an uninoculated part of the plate.

  7. Sterilise and quench the loop again.

  8. Streak from the secondary inoculation toward an uninoculated area of the plate, being careful not to contact the primary inoculation, or the culture will not be properly diluted.

  9. Repeat 6-8 a couple more times, ending in a squiggle in the centre of the plate.

The result of this is that bacteria are eventually deposited individually across the plate with sufficient separation that after they are incubated, the colonies are separate, and can be considered to be a single species.

Then, we have the techniques needed to isolate potential antibiotics. Once isolated, the potential antibiotics are impregnated into standard-sized absorbent disks in known concentrations and placed on a lawn of the target bacterium. When cultured, the efficacy of the antibiotic can be shown by the size of the clear area around the disk that carries it. Without knowing concentrations, you may reject a better antibiotic just because you didn't collect as much of it.

How long would it take our microbiologically ignorant time traveller to re-create even these basic techniques?


Isolating a fungus that produces an effective and safe antibiotic is the next problem. Fleming discovered Penicillin effectively by accident, and it took around ten years before he and his successors Florey and Chain succeeded in isolating a strain of penicillium mould that produced penicillin in sufficient quantities to be commercially viable. "Mouldy" Mary Hunt, Florey's assistant, spent much of her time scouring local markets for mouldy produce in the hopes of finding an active penicillin-producing fungus. Our time traveller should be grateful to find any antibiotic-producing fungus, let alone an active penicillin-producing strain.

A great many fungi do produce bactericidal compounds, but these compounds are either themselves - or are accompanied by - compounds toxic to humans. There is little point administering an antibiotic which will also sicken or kill the patient. The challenge is in finding one with sufficient specificity that produces its active compound in sufficient quantity.


Antibiotics tend to be specific to certain bacteria. Penicillin is specific to mostly Gram-Positive bacteria, and treating a patient with a Gram-Negative bacterial infection may be completely ineffective. Whatever antibiotic our time-traveller is lucky enough to discover could well be similarly specific.


Once a suitable strain is found, it must be produced in sufficient quantity. Industrial fermentation tanks have their own special engineering and microbiological challenges, such as cleanliness, temperature regulation (Cooling can be as important as heating), nutrient supply, oxygen supply, foaming control, agitation and harvesting. Operation of industrial fermentation equipment is a part of the coursework for later years of study in Microbiology that I never undertook - so obviously not a trivial subject.

Potential results of proposed approach

If our ignorant time traveller feeds already ill patients with various moulds or fungi in the hopes of chancing upon one that produces an antibiotic, then he is doomed to frustration and failure. The odds are that few fungi produce suitable antibiotics, and none are likely to produce sufficient antibiotics to have any effect. Most likely, his patients will die or recover on the strengths of their own immune systems, and their tolerance to both the initial bacterial infection and the fungal infection that our well-meaning but ignorant time traveller subjects them to.

Finding a patron rich enough to fund all these experiments - whether conducted properly as I have detailed, or otherwise as the OP detailed - would be quite difficult. The most likely patron would be royalty or nobility.

Then, the second difficulty that our aspiring microbiologist would face would to remain unexecuted and unimprisoned in the face of an impatient nobleman without the least idea of what he signed up for demanding results now, or else! when the experiments drag on apparently interminably without the least signs of results other than piles of glassware and other strange apparatus, and ongoing bills for various strange and difficult-to-procure compounds.

Even a noble patron with the patience of a saint could make all sorts of inappropriate suggestions for speeding matters and for cost-cutting, and may intercept and alter purchase orders, in a well-meaning way, of course. ("Why does he need agar? Let's substitute gelatine, it's cheaper!") Our time traveller would have to educate his patron (or a trusted advisor) nearly as well as he was himself in order to be fully understood and properly supported.

If our time-traveller succeeded at all, he'd likely be an old man by the time he did so.

  • 4
    $\begingroup$ Thank you for your answer, I really appreciate it and find it informative, thus my voting for it. However, I did notice that your focus is on how hard it is to do purely rigous process of today, which is true. However, Flemming and the others did accomplish the feet without todays tools. It would be nice to focus a little more on viability on less-then-perfect approach and how well it can work. For instance, could the traveler find bacteria to test molds on without growing it himself somehow? Can rigorousness of today be sacrificed and still get some sort of result? $\endgroup$
    – dsollen
    Commented Sep 2, 2015 at 17:00
  • $\begingroup$ for instance, moldy mary approach of finding fungus seems viable in any timeframe. If the traveler realized early on that finding it is easier then growing it himself and used a similar approach he should have a success in roughly the same time as mary right? if he has a larger backer and multiple people working to hunt mold perhaps even a faster success? It may even be easier to find mold in medieval times because of worse food storage techniques. I know it's not up to today science standards, but it's better then certain failure if today rigor is doomed to fail due to cost and time. $\endgroup$
    – dsollen
    Commented Sep 2, 2015 at 17:05
  • 1
    $\begingroup$ @dsollen, without rigor, you can't know that you have an effective antibiotic, and you can't isolate the most effective strain of fungus. The problem with having more assistants is that unless some are trained to do the microbiological work, you can't process what your field assistants bring you fast enough. An antibiotic is only effective in humans in truly vast quantities compared to what the fungi produces, so you would have to produce lots of it. In the medieval period, these would be nearly insurmountable problems. $\endgroup$
    – Monty Wild
    Commented Sep 3, 2015 at 1:02
  • 1
    $\begingroup$ Btw, Fleming, Florey et al used the basic techniques I mentioned here. These days, there are machines that do all of this in a fraction of the time, which would obviously not be available in the medieval period. $\endgroup$
    – Monty Wild
    Commented Sep 3, 2015 at 1:05
  • 3
    $\begingroup$ @dsollen "However, Flemming and the others did accomplish the feet without todays tools." No, he just had 1928 cutting edge tecnology, 30+ years of experience in the field and was surrounded by experts and brilliant lab helpers. In the other hand, you have a layman with medieval technology, surrounded by experts for whom it would be a huge step to believe in phlogiston (forget about germs), untrained (probably illiterate) helpers and a level of knowledge in the field summarized as "it was something about bread and a fungus".... $\endgroup$
    – xDaizu
    Commented Jul 4, 2017 at 9:59

There are a number of faults with your premise:

Availability of Penicillin

Penicillium is actually a very common genus of fungi, and grows all over the place. Flemming's accidental discovery of penicillin wouldn't have happened otherwise. However, penicillium won't just produce penicillin for the fun of it. It only produces penicillin when under stress (under attack by bacteria). Simply growing the fungus on bread won't necessarily get you penicillin.

Bacteria and Petri dishes

Petri dishes aren't anything special. You could use bowls, plates, etc for the same effect. The key ingredient here is agar, which is the growth medium used to create large macroscopic bacteria colonies. Spitting into a metal bowl won't get you what you want.

And that's going to be the biggest problem: growing colonies of bacteria, that are just bacteria. Unless our time traveler can create a sterile, viable growth medium, then he's going to have a tremendously difficult time isolating bacteria from fungi (which will likely be some kind of antibiotic producing fungi at that).


Our time traveler is likely going back in time at least a few centuries before the age of reason. People have no understanding of germ theory, the scientific method, or modern medicine. He might be able to get a few monks on his side, but he'll have a difficult time convincing a king that:

  1. He's not a raving lunatic.
  2. He needs to go through so much ridiculous bureaucratic nonsense to make a cure.
  3. He's worth the funding.

At the end of the day, we have to ask the correct question: Is the knowledge that the blue-green fungus that grows on fruits and bread produces a antibiotic compound enough to produce an effective antibiotic?

I believe that the answer is, yes. However, it won't be through a complicated trial-and-error, double-blind experiment type of method. It will simply be by feeding people mold. Eating penicillin is an effective way to get the antibiotic into the body. Also, you can just slap some mold on cuts or gashes, and that will help keep the wounds free from infection.

The most valuable skill, at the end of the day, is being able to identify penicillin in the wild. There are some other bad molds and fungi out there, so simply knowing that penicillin is a mold isn't as useful as it could be otherwise.

  • $\begingroup$ so if the traveler spits into a petri dish with lots of sugar (or other 'food', he could experiment with different types) and closes it with glass or other method he will not be able to ensure bacteria is the primary growth? It seems that if sterilized beforehand, only getting spit from someone, and then cutting it off from other sources would be enough to usually grow bacteria, it's okay if a few bad dishes exist since he would be trying many. Could he not learn to tell the difference between bacteria and fungus visually? $\endgroup$
    – dsollen
    Commented Apr 1, 2015 at 16:37
  • $\begingroup$ @dsollen How do you sterilize it? Use your pressure boiler? What about using sterile latex gloves while handling it? How would you incubate it at a highly controlled temperature? My point is that the effort our wayward traveler would have to undergo to properly test it according to medical standards of today would be useless because such a test wouldn't actually give you that much. The number of variables involved contributes to far too much error. Simply dosing people with mold would 1) be much simpler, 2) give you as much information, and 3) significantly less ridiculous to the locals. $\endgroup$
    – Nick2253
    Commented Apr 1, 2015 at 16:42
  • $\begingroup$ make a metal dish and throw it in a hot fire for awhile. Either let it cool naturally with a 'lid' on it, or cool it rapidly using boiled water (try both and see what works). It wouldn't be near perfect, but I would think it would work 'well enough' for imperfect trials. And yes I understand that testing fungus may seem easier, but if our traveler is unsure of what he is looking for, or potential harm of screwing up, he would want to do things carefully. I'm working on the assumption that he impressed the king with future tech so he gets a bit of leeway with his approach, he did 'magic'!! $\endgroup$
    – dsollen
    Commented Apr 1, 2015 at 16:46
  • 1
    $\begingroup$ Sigh. I've studied university-level microbiology, and the problem isn't as simple as this suggests. The first person treated with penicillin died because they couldn't get enough into him, even though they were distilling the penicillin out of his urine and giving it back to him. Just because a fungus produces an antibiotic doesn't mean that it will produce enough, or that it will be safe for humans. It took Fleming's assistant 'Mouldy' Mary years to find a more active strain. If you just eat mould, you could end up with a fungal infection on top of the bacterial infection. $\endgroup$
    – Monty Wild
    Commented Aug 26, 2015 at 3:18
  • 1
    $\begingroup$ @dsollen, Done. Have a look at my answer. $\endgroup$
    – Monty Wild
    Commented Sep 2, 2015 at 1:20

This approach seems sound, and may be the proper approach given what the time traveler knows and the tools available. (It also appears to be a similar approach to the one taken in this paper) After all, penicillin was discovered by some contamination. There are some issues:

  1. Penicillin has been found to grow best on corn, which is a new world crop, and cantaloupe, which does not natively grow in Europe. Finding a suitable substrate for growing will be just as hard as getting the correct mold. (That being said, cheese and sausages may be the time traveller's best bet.)
  2. Penicillium Molds can produce mycotoxins, toxins from fungi, which can be deadly to humans. Only certain species can produce penicillin; others may kill off bacteria with their mycotoxins. The time traveller will need to be very careful in selecting the moulds.

I would say that the time traveller would expect to see more negative results than positives. The time traveller is looking for a particular kind of mold, and even molds within the same family may not have the desired effect. Although penicillium molds are everywhere, the correct ones may not be captured.

This effort could take his lifetime, but likely on the order of years. Finding a suitable substrate to grow the mold on for mass production may never happen, because of #1 above. I see it being an uncommon, but potent, cure. Therefore richer families (or those whom the time traveller takes pity on) would profit from this.

  • $\begingroup$ Also consider that most noble patrons want quick results or at least visible progress, and they won't get them from this project. $\endgroup$
    – Monty Wild
    Commented Aug 26, 2015 at 3:20

You must log in to answer this question.

Not the answer you're looking for? Browse other questions tagged .