16
$\begingroup$

I am a fictional writer and need info on a poison that can be put in food and is undetectable in autopsy. Symptoms also can't be too messy, no specific time in how long it takes to take effect.

$\endgroup$
18
  • 7
    $\begingroup$ Do you need to avoid detection entirely ("no idea why he died, he seems perfectly healthy!") or would making it look like an accident ("his drinking habit finally did him in") be sufficient? $\endgroup$
    – Draconis
    Dec 31, 2016 at 15:59
  • 73
    $\begingroup$ Fictional writer asking questions on WB would be a great cover for a murderer doing research. $\endgroup$ Dec 31, 2016 at 16:18
  • 34
    $\begingroup$ Sometimes I seriously wonder if some of the people on this site are on an FBI watch list. $\endgroup$ Dec 31, 2016 at 17:26
  • 28
    $\begingroup$ No mention of Iocane powder yet? It is odorless, tasteless, dissolves instantly in liquid, and is among the more deadlier poisons known to man. $\endgroup$ Jan 1, 2017 at 0:15
  • 8
    $\begingroup$ Okay, wait a minute. You say you're a fictional writer. Why should we explain how to kill someone undetected to someone who is only pretending to be a writer? $\endgroup$ Jan 1, 2017 at 21:34

18 Answers 18

34
$\begingroup$

Science and autopsies have reached a point in the modern world where poisons are generally detectable.

The best poisons are ones that break down into elements that occur naturally: succinylcholine(SUX) and potassium chloride. SUX causes asphyxiation and paralysis, however it is a very painful death. Potassium chloride causes severe heart arrhythmias and mimics a heart attack.

These break down to elements natural in the body and would easily be overlooked: succinic acid and choline for SUX, potassium and chloride of course for potassium chloride which is common in heart attack victims due to muscle damage. However both need to be injected and leave an injection site, this may be pesky if your victim does not take any injected medicine or drugs. While these are the least detectable they don't meet the criteria of being ingestable.

Aconite is another great one. It leaves Aconitum alkaloids, but these can only be detected by a gas chromatography/mass spectrometry. Death usually results from paralysis of the respiratory system or cardiac arrest. What is useful about this poison is it can be absorbed through the skin or consumed and only takes 2mg of pure aconite or one gram of the plant to kill. This method is fairly undetectable as long as there is no cause for an in-depth autopsy and it can be ingested.

Depending on your victims history, there are other ways to poison them and make it look accidental, for example if your victim loves blowfish, you could poison them with tetrodotoxin which naturally occurs but is generally removed by a skilled cook in the cooking process.

$\endgroup$
6
  • $\begingroup$ Could I have a micro-robot that fits inside an ordinary pill and injects the poison from inside when swallowed, then dissolves in hydrochloric acid? $\endgroup$ Jan 1, 2017 at 12:04
  • $\begingroup$ @JanDvorak, I would need more info for a complete answer which would be better handled in a posted question. Short answer is not with today's technology. In the future we may be able to develop sufficient nanobot tech $\endgroup$ Jan 1, 2017 at 14:41
  • $\begingroup$ What about an insulin overdose? easy enough to send somebody into a diabetic coma... if they have an implant, it most assuredly can be hacked to temporarily up dosage... $\endgroup$ Jun 4, 2017 at 23:32
  • $\begingroup$ @IsaacKotlicky In the past actually insulin used to be a perfect poison. I remember being told by my profesor that nowadays there is an indirect way of detecting that. $\endgroup$
    – Shadow1024
    Jun 5, 2017 at 14:58
  • $\begingroup$ @IsaacKotlicky Insulin did not match the OPs request that it could be put in food and is undetectable. Insulin overdose or the inverse glucose overdose can also be difficult to use to kill a person, some peoples bodies can handle much more than others before their bodies go into shock. Getting the right dose without making it obvious would be hard unless you are a doctor. $\endgroup$ Jun 5, 2017 at 18:21
15
$\begingroup$

Not directly

There is nothing that can kill you directly that an autopsy could not find eventually, your best bet is to use something that is circumstantially deadly but would not look out of place: for instance salamander toxin in a hikers cookware. It just looks like a sad accident.

Swapping prescription medication this works even better if they have risky meds, just switch the contents of the bottles around, lots of people die becasue they do this to themselves.

Deathcap mushrooms mixed into the salad of a home gardener. That one kills a lot of people already. Mostly becasue the symptoms take days to set in, look like the flu at first, and the mushroom looks like the edible ones and tastes quite good.

Lots of people fatally poison themselves all the time just pick an appropriate one.

$\endgroup$
4
  • $\begingroup$ Did you mean Salmonella ? Or do Salamanders have a poisonous toxin? $\endgroup$
    – Criggie
    Jan 1, 2017 at 9:12
  • 3
    $\begingroup$ @Criggie Some salamanders are poisonous. Just like a lot of other amphibians that are on the menu-list of other animals (of frogs and toads it is more well-known). $\endgroup$
    – Tonny
    Jan 1, 2017 at 13:07
  • 5
    $\begingroup$ A common way of poisoning in hikers is various toxic salamanders that will crawl into kettles or stew pots then get boiled unbeknownst to the hiker, who then dies from ingesting the toxins in the water. I know both california and europe both have such salamander cases. $\endgroup$
    – John
    Jan 1, 2017 at 16:19
  • $\begingroup$ @tonny and john I have learned something new today - thank you! $\endgroup$
    – Criggie
    Jan 1, 2017 at 23:13
10
$\begingroup$

I'm sure people remember the poisoning of Alexander Litvinenko with a Polonium isotope? Wikipedia says that only a random coincidence allowed the cause of death to be discovered. But from now on I'll bet that in high-profile autopsies, alpha and beta emitters will be tested for as cause of death.

But, I do think the side-effects can be messy...

$\endgroup$
1
  • 3
    $\begingroup$ Polonium is a bit expensive and hard to buy. In this case I even wonder, whether it wasn't picked on purpose to make clear who was behind the poisoning. $\endgroup$
    – Shadow1024
    Jun 5, 2017 at 14:55
4
$\begingroup$

Victims of oxygen-deficient atmosphere pass out quickly and die. If a victim was to breathe pure nitrogen, say, for a while, he would die. There would be no way to test for nitrogen since it is 70% of the atmosphere and very common in the body. The deceased would have very little oxygen in their blood or tissues, but I imagine this is common to all dead people.

Sorry, I know it's not a food poision, but maybe it will help.

See also the book The Poisioner's Handbook. But, it's mostly about how poisions can be detected.

$\endgroup$
2
  • 1
    $\begingroup$ I'm no doctor but I'm fairly sure it should be easy to tell they died from asphyxiation, even if they don't know which gas was used $\endgroup$ Jun 5, 2017 at 2:50
  • 3
    $\begingroup$ Hypoxia causes a fairly distinctive type of brain damage which can be observed at autopsy. It won't reveal the cause of the oxygen deprivation, but it will certainly reveal it happened. $\endgroup$ Jan 2, 2018 at 22:06
4
$\begingroup$

I can't think of anything specific, but as has been said, poisons are detectable.

This means that you need to hide it. One way to do that, as has been said is to make the poison degrade into another chemical. However, the remains will still be there.

So I recommend hiding in plain sight:

Use a chemical that is needed by the body, but use a double or triple dose. I remember seeing an episode of some detective series, where someone expresses regret for killing someone by giving them an overdose of a certain vitamin. I don't know if that would be possible, and the chemical would still be there, but people probably won't look for clues in the amount of certain necessary vitamins and minerals.

Another thought I just had is Chirality

A lot of organic molecules have two distinct forms which look the same on paper, but one is needed and used by the body, the other is harmful. These are called the "right" and "left-handed" versions. I once read about a medicine (EDIT: the medicine is Thalidomide (thanks @Nikolai)) released in Germany pre-world-war-II, that caused more harm than good, because they used the wrong handed molecule, well actually both, but the wrong molecule caused quite a few stillbirths if I remember correctly (the medicine was given to pregnant women). I think I remember that humans, and all other life on Earth uses left-handed molecules, so if you give someone right-handed carrots or potatoes, then you could poison them quite badly.

Since the two chemicals are almost the same, it would be undetectable.

$\endgroup$
9
  • 4
    $\begingroup$ Vitamin C overdose sounds like nonsense; it has unbelievably high LD50 compared to other normally-ingested things. $\endgroup$ Jan 1, 2017 at 6:22
  • $\begingroup$ @R.. it was a long time ago. $\endgroup$ Jan 1, 2017 at 15:13
  • 2
    $\begingroup$ the word it's chirality $\endgroup$ Jan 1, 2017 at 15:34
  • $\begingroup$ The drug you're referring too is called Thalidomide, It was designed as pain medication (and was therefore used in pregnant women), nowadays its used as an anti-cancer drug. $\endgroup$
    – Nicolai
    Jun 4, 2017 at 13:56
  • 1
    $\begingroup$ Wow, is there actually any way of growing right-handed carrots? $\endgroup$
    – wleightond
    Jun 6, 2017 at 11:20
2
$\begingroup$

Take something with a half life of at most a day and whose decay product normally exists in the body. Death by radiation exposure, it's decayed to effective zero by the time of autopsy. I haven't checked all the radioisotopes to see if such a beast exists, though.

$\endgroup$
3
  • 7
    $\begingroup$ Problem is, acute radiation poisoning tends to be quite obvious. And rather messy. Skin-falling-off messy. $\endgroup$
    – Bob
    Jan 1, 2017 at 6:20
  • $\begingroup$ @Bob He didn't ask for the effects not to be obvious. $\endgroup$ Jan 1, 2017 at 20:47
  • 3
    $\begingroup$ "undetectable in autopsy. Symptoms also can't be too messy" -- it's like half the question. If it's obvious, then it's easy to guess what the cause is, then it's not really undetectable even if the exact agent has dispersed (or decayed in this case). To me, undetectable implies that it appears like a natural death (or a disease rather than an assassination). $\endgroup$
    – Bob
    Jan 2, 2017 at 0:17
2
$\begingroup$

Heavy water, D2O, would kill your victim within a week if you could substitute it for most of the H2O they would normally ingest over that time.

It could be detected, but only via mass spectrometry or gas chromatography. Tests for both common and uncommon poisons would produce negative results.

It would be a somewhat expensive way of killing someone though as in purified form it sells for around $1200/litre.

$\endgroup$
1
  • $\begingroup$ Ah. Apologies Penguino, I apparently duplicated your answer accidentally. I answered similarly, and after you, but did not see your answer in the list. $\endgroup$
    – PcMan
    Feb 23, 2021 at 13:33
1
$\begingroup$

A bit far fetch, but if you swallow a large quantity of dry ice (solid state CO2), you would probably suffocate.

An autopsy would show a suffocation due to excess of CO2, but not that it came from the stomach

$\endgroup$
2
  • 4
    $\begingroup$ You would have to consume a large amount of dry ice for it to kill you and from what I can tell the cause of death will be one of two things: freezing your internal organs or, more likely, excess bloating as the gas expands. Just because it is in your stomach does not mean it will get in your lungs. For suffocation you would have to be in a room filled with CO2 or place a bag filled with it over your head and breath deep. $\endgroup$ Dec 31, 2016 at 21:49
  • 3
    $\begingroup$ Also, the stomach is likely to have some kind of internal frostbite, I imagine. $\endgroup$ Jan 1, 2017 at 10:27
1
$\begingroup$

Detection is easy enough with mass spectrography these days for blood and tissue analysis.

Insulin is always a good one as the victim just slips into a coma and, with enough insulin, it stops the brain from functioning, long enough in that state and someone dies. Of course if someone finds the person it's very easy for a paramedic to find out they are hypoglycemic and all they need to do is give them a sugar-saline drip.

As insulin is common in the body it used to be missed, but all autopsies now look at all chemical levels of the common types found in blood to see if anything was amiss.

If the victim is a hippy and has not had vaccinations then you could use tetanus bacteria and paralyze them.

Another way would be to use bacteria that can cause meningococcal, then somehow administer it into their eyes (thus straight into the brain), hell if the victim uses eyedrops or contact lenses you can get them to administer it themselves.

$\endgroup$
1
$\begingroup$

Botulism toxin has medical applications and works in such trace amounts (basically shutting nerves down without being used up in the process) that the principal detection method is trying to replicate symptoms in mice. However, this requires either access to the original ingested substance or significant overdosage (so that viable samples may be extracted from the body of the victim). In addition, it requires a solid hunch regarding the cause of death (nominally asphyxiation): you don't test for it without reason.

I consider it likely that medical supplies are tagged with marker substances but it's not all that hard to cultivate on your own.

$\endgroup$
1
$\begingroup$

Use Ricin, or, even better, a similar protein.

Ricin is a toxic protein produced by Ricinus communis. It is toxic if ingested and therefore does not leave any injection mark.

It has the effect of blocking protein synthesis and is therefore toxic to every cell type. Depending on doses it can kill in hours or a few days.

It's historically been used for assassiantion and therefore its synptoms are well known, however:

Synptoms may vary from person to person, also, it is possible that a purer for of the protein may have fewer symptoms

there are no common clinical tests for Ricin poisoning, bu it can be detected in 2 ways, DNA amplification and antibodies essays.

What to do then? Use a different protein. There are many protein similar to Ricin, Abrin for example is lethal at even lower concentration which makes detection harder; even better you can DESIGN a different protein. Use genetic engineering to remove the part of the protein that binds to the antibodies so that it won't be detected with antibodies essays and instead of extracting it from the beans express it in a vector and then purify it(in this way the DNA detection won't find any traces)

$\endgroup$
1
  • $\begingroup$ there are no common clinical tests for Ricin poisoning in use today. But there are plenty of candidates (including the antibody assays you mention, plus a few more recent approaches that are more easily mass-produced) and interest in putting them into use due to a number of high profile incidents recently, so within the next 5-10 years I imagine you will start to see Ricin testing as a common test when appropriate symptoms are exhibited and other more common causes are ruled out. $\endgroup$ Jan 1, 2017 at 16:16
1
$\begingroup$

As other answers have suggested, poisons that can't be detected are basically a thing of the past in the modern world. Pretty much anything that has ever been identified as a poison can be tested for, and if there is a reason to do so it will be tested for. Your only hope of having the poison undetected, therefore, is giving the investigators a reason to stop looking: a plausible explanation for why the victim died that can be chalked up to accidental circumstances.

I don't know what the circumstances of your plot require, but one possible approach would be rather than have your victim poisoned you could have the perpetrator drug them instead. This is particularly possible without looking like an issue if the victim regularly uses sleeping tablets, or is a recreational depressant user (e.g. opiates like heroine or opioids like codeine). Then, when they're insensible, inject a massive dose of alcohol. A cursory examination will show that they've gone to sleep under the influence of too much drink and drugs that react badly with it, and died of the consequent alcohol poisoning. Verdict: death by misadventure, investigation closed.

$\endgroup$
1
  • $\begingroup$ Observation from a case here: A "perfect" murder was solved because the needle puncture was found. (They were suspicious, though.) $\endgroup$ Feb 22, 2021 at 23:18
1
$\begingroup$
  1. Salt.

Drinking concentrated salt solutions was a method of suicide / execution in China. http://www.nepachemistry.com/2010/12/death-by-common-salt-nacl.html

Our bodies are naturally full of salt. After death, fluids continue to equilibrate by mass action, with salt moving off into tissues and water moving in from tissues to hydrate the salt. It would be difficult to establish that salt was the culprit if any time had passed after death.

  1. Water. Overdose of plain water can kill. It has happened more than once with water drinking contests. https://www.scientificamerican.com/article/strange-but-true-drinking-too-much-water-can-kill/ The water dilutes out needed electrolytes and causes a heart arrhythmia. After death the water and body salt equilibrate and as with the opposite circumstance above, it is tough to prove what happened.

Technically this meets the OP: can be put in food and undetectable at autopsy. Neither water nor salt has a lot of promise for surreptitiously poisoning someone. Although if you encapsulated the salt somehow such that it was not tasted that might do it.

$\endgroup$
1
  • $\begingroup$ Certain club drugs, e.g. ecstacy a.k.a. Molly a.k.a. MDMA), can induce people to drink water in amount that can be deadly. While it can be detected in a person's system, if the person were a clubbing regular, the presence in the system might not be connected to the death, particularly is the water they sought was replaced with saline (i.e. salt water), exacerbating rather than mitigating the dehydration effect from the MDMA. The MDMA could be put in the food trigging the eventual death by saline. $\endgroup$
    – ohwilleke
    Aug 20, 2018 at 17:34
1
$\begingroup$

You might notice the sting of the injection. Within seconds you'd realize you're having trouble moving your eyes and fingers, followed by your arms and legs. If you were standing, you'd collapse. In a heap on the floor, you'd realize nearly every muscle in your body was paralyzed. Being fully conscious, your sense of panic would be rising as rapidly as the paralysis was spreading. Swallowing and breathing has become more and more difficult. Slipping into unconsciousness, your last conscious thought may well be "I am going to die."

From Gizmodo: The History of Sux, the World's Most Discreet Murder Weapon
Read the whole article there.

SUX is a posion that would be good for a crime scene. It is detectable, yes, but it's very difficult to detect. Might be an idea

$\endgroup$
1
  • $\begingroup$ Welcome to WorldBuilding! Interesting answer. If you have a moment please take the tour and visit the help center to learn more about the site. Have fun! $\endgroup$
    – Secespitus
    Jan 1, 2018 at 12:27
0
$\begingroup$

Pretty much anything can be detected if the detective knows what to look for. A killer's best shot is using a poison that is very rarely used and hence would not be looked for. Succinylcholine is a drug pretty hard to trace. It paralyzes the muscles including those that we use to breathe.

$\endgroup$
5
  • $\begingroup$ suffixation symptoms are pretty obvious $\endgroup$
    – MolbOrg
    Dec 31, 2016 at 18:12
  • $\begingroup$ Yes but it is not obvious that the poison caused them. $\endgroup$ Dec 31, 2016 at 20:38
  • 1
    $\begingroup$ yeah at the start, but without other obvious reasons water in lungs, hands prints, pillow nearby etc - everything boils down to the expertise of the autopsist or medical examiner (depending on the freshness of the object) - that's my guesses, not competent enough to say $\endgroup$
    – MolbOrg
    Dec 31, 2016 at 22:49
  • $\begingroup$ @MolbOrg What's obvious about them? Suffocation by force leaves signs, that says nothing about an anoxic atmosphere or a simple failure to breathe. $\endgroup$ Jan 1, 2018 at 21:03
  • $\begingroup$ @LorenPechtel obvious that they have to have a cause. Those are external conditions or medical conditions. I'm over my head there and kinda admitted that by my second comment, and your comment is reasonable, but disability of detecting reason of suffocation, assuming fresh enough body, I guess it would be strange. Symptoms are obvious as long as heart works, it is visible because arterial and venous blood has a different color, and as I heard, a body will have a different color. And probably that is what I meant by obvious, typical color of the surface tissues of the dead body. $\endgroup$
    – MolbOrg
    Jan 3, 2018 at 10:00
0
$\begingroup$

As working in cardiology, I see frequent mistaken ODs with high doses of Cardiology drugs Digitalis is frequent. However, Beta Blockers at high doses, Calcium Channel Blockers with Negative Inotropic effects can cause "Heart Block" and thus heart Failure. This Scenario was suggested in the TV series "DEXTER" for his step father, Harry. Many people OD themselves on their prescribed medications by accident verses purposely, but most often EMTs are CALLED to bring them in by a relative or friend? So, the case would need to be done in a more remote area as happens with backpackers or wilderness campers. The issue is what would the victim with a Heart ailment be doing out in a wilderness area and what would the authorities think of the companion survivor. However, it happens with WIFE survivors with their husbands frequently on vacations without suspect? Not always so, the other way around without suspect--ha ha! Try an angry wife villain with good assets but not a large Life Insurance? Rich Husband OK, Rich wife not so good ha,ha! Although Divorce works well for women without physical murder for obtaining assets. Mrs Scarlett did it with Atenolol in the RV??

$\endgroup$
2
  • $\begingroup$ Welcome to Worldbuilding! Check the tour. Note that it's better to just answer, and not add diversionary chatty text. A couple paragraph breaks would be good here, too. $\endgroup$
    – JDługosz
    Jun 4, 2017 at 10:12
  • $\begingroup$ Thinking about ER admits, people taking high dose antibiotics purchased from a vet or pet store can lead to an ER visit and potentially kill, but a medical examiner would probably think it was an OD caused by not being able to afford to get a prescription and human meds, rather than a murder attempt. $\endgroup$
    – ohwilleke
    Aug 20, 2018 at 17:39
0
$\begingroup$

Grapefruit juice

In order for this to be effective, the victim must be on medication that must be swallowed.

There is a whole wiki article about it, I'll just leave the juicy bits (drumroll) here.

One whole grapefruit, or a small glass (200 mL, 6.8 US fl oz) of grapefruit juice, can cause drug overdose toxicity. Fruit consumed three days before the medicine can still have an effect. The relative risks of different types of citrus fruit have not been systematically studied. Affected drugs typically have an auxiliary label saying "Do not take with grapefruit" on the container, and the interaction is elaborated upon in the package insert. People are also advised to ask their physician or pharmacist about drug interactions.

The effects are caused by furanocoumarins (and, to a lesser extent, flavonoids). These chemicals inhibit key drug metabolizing enzymes, such as cytochrome P450 3A4 (CYP3A4). CYP3A4 is a metabolizing enzyme for almost 50% of drugs, and is found in the liver and small intestinal epithelial cells. As a result, many drugs are affected. Inhibition of enzymes can have two different effects, depending on whether the drug is either

  1. metabolized by the enzyme to an inactive metabolite, or
  2. activated by the enzyme to an active metabolite.

In the first instance, inhibition of drug-metabolizing enzymes results in elevated concentrations of an active drug in the body, which may cause adverse effects. Conversely, if the medication is a prodrug, it needs to be metabolised to be converted to the active drug. Compromising its metabolism lowers concentrations of the active drug, reducing its therapeutic effect, and risking therapeutic failure.

Low drug concentrations can also be caused when the fruit suppresses drug absorption from the intestine.

So either the victim is affected because their medication stops working, or they overdose it.

If you like infotainment videos, here is one where a youtuber describes some cases of grapefruit induced overdose. In a couple cases those were fatal.

$\endgroup$
0
$\begingroup$

Nothing is undetectable in a dedicated autopsy.

But how about something that is very, very, very unlikely to be tested for?

Replace your victim's water supply with Heavy Water.

It looks like water. It tastes like water. It satisfies your thirst like water. It is not any more or less radioactive than water. It even has boiling and freezing points so close to normal water that the difference is invisible.

But, if you replace enough of the normal water in your body with heavy water, you die. With symptoms that resemble systemic poisoning by a cytotoxin, or highlevel radiation poisoning.
Like the bite of a brown recluse spider, for example. Except of course there would be no bite site to be found.
Nor, if radiation is suspected, would a radiation counter be at all interested in the victim. D2O, after all, is quite inert.

While it is quite possible to test for heavy water, no normal autopsy would even begin to hint to the need for this. It would show on a mass spectroscopy test, except that the lines for water are routinely ignored in these tests, due to all biological sample being stuffed to the gills with water.

$\endgroup$

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