Measuring Deadliness | Toxinology 101

By Christie Wilcox | April 30, 2017 7:00 pm

Scientists refer to the study of biological toxins as toxinology. From bacterial toxins like anthrax to the deadliest snake venoms, toxinology examines the chemical warfare between animals, plants, fungi and bacteria. In my Toxinology 101 series, I explain and explore the fundamentals of toxin science to reveal the unusual, often unfamiliar, and unnerving world created by our planet’s most notorious biochemists.


One of the most frequent questions I receive as a venom scientist (so much so I dedicated an entire chapter of my book, Venomous, to it) is some variant of What is the deadliest toxic animal? While that seems like there should be an easy answer, as with anything in the natural world, defining deadliness is messy. To answer that question, you have to be clear about what you’re really asking. Is the subtext of the question What animal is most likely to kill me? Or What animal should I be most afraid of running intoOr more simply, What animal produces the most potent toxin, because I’m a biochemistry nerd and I’m just curious? Each of those questions is answered a bit differently, and even still, it’s complicated.

Mirror mirror on the wall: What toxin is the deadliest of them all?

Let’s say you really just want to know what species produces the deadliest toxin from a biochemical standpoint. You don’t care about calculating your personal risk of dying in any way, you just want to know what toxin packs the worst punch. If you had a vial containing 1 gram of the most potent toxin produced by a living creature, which toxin would it be?

The most common measure of potency used by toxinologists is what is called a median lethal dose, or LD50 for short. This is the dose per body weight that kills 50% of a group of animals that receive it. This value can be calculated in a lab, and therefore in theory allows for standardized comparisons of the deadliness of different toxins and toxic mixtures. But of course, in practice, even this is much messier than it sounds.

First off, LD50 values haven’t been calculated for most toxic species. But even those that have been calculated can vary substantially dependent on several factors. The first is how the toxin is delivered, as toxins can act very differently when administered to different parts of the body. Scientists can feed toxins to lab animals (orally), or they can inject the toxins into the creatures. And when they do the latter, they can inject them just under the skin (subcutaneously), deeper into the body tissues (intraperitoneally), or directly into the bloodstream (intravenously). Each of these routes of administration may have different LD50 values.

Tail vein injections are one of the most common methods for intravenous injection in mice. Photo by Armin Kübelbeck,Wikimedia Commons

Tail vein injections are one of the most common methods for intravenous injection in mice. Photo credit: Armin Kübelbeck/Wikimedia Commons

To demonstrate this, let’s take a look at a relatively recent case study. Back in 2015, scientists announced the discovery of the first venomous frogs. Two species of frog, they revealed, possess potent toxins that are stabbed into potential predators by bony protrusions of the animal’s skeleton. News articles describing the discovery often contained impressive descriptions of the frogs’ deadliness. “Tropical amphibians thrust out bony spines to deliver toxin capable of killing 80 people,” touted the headline from The Daily Mail. “World’s first venomous frog has the kiss of death,” claimed New Scientist. Or, according to NBC News,First known venomous frogs could kill you with a head butt.”

All of these sensational headlines are derived from the LD50 of the frogs’ venoms. The two species were found to have pretty potent toxins. Aparasphenodon brunoi, the deadlier of the two venomous frogs, had intraperitoneal LD5os of  3.12 μg per mouse for venom from the head. Since that dose was in mice that ranged from 18 to 20 grams in weight, we can calculate a standardized per victim body weight LD50 of 0.16 to 0.24 mg/kg for the head venom from A. brunoi. The less deadly species, Corythomantis greeningi, had LD50s ranging from 2.5 to 2.9 mg/kg. If you scale that dose up to the average human, that means 1 gram of the most potent venom from A. brunoi could theoretically kill 80 people—hence the scary language in the media reports.

The two venomous frogs and their spiny skulls which are likely used to envenomate would-be attackers. Figure 1 from Jared et al. 2015

The two venomous frogs and their spiny, venomous skeletal projections. Figure 1 from Jared et al. 2015

But there’s a catch. The paper describes the animals as possessing tiny little bony points on their skeleton which deliver the venom, each of which is less than a few millimeters long. That means that there’s simply no way that the venom from these frogs is going to be delivered deep into the body, intraperitoneally, as the LD50 tests were run. Instead, at best, the venoms would be delivered just under the outer layers of skin, subcutaneously. The paper didn’t calculate a subcutaneous LD50, but it did run another experiment which gives us a clue as to what the subcutaneous LD50s might be. To study whether the venom induced swelling (what doctors call edema), the researchers injected varying doses of venom into the fleshy paws of mice (subcutaneously) and monitored the area for three days. For the most potent venom, those doses ranged from 0.125 to 32 μg. Remember, the dose that killed half of the mice when they injected it intraperitoneally was 3.12 μg. Since none of the mice injected subcutaneously died, we can conclude that the subcutaneous LD50 of the venom is more than ten times less potent than the intraperitoneal LD50. We can also conclude that given such a large subcutaneous LD50 (>1.8 mg/kg, or roughly 150 mg of toxin in a single person), there’s pretty much no way these little frogs could kill a human being, even with multiple head butts.

Of mice and men

Even if these frogs could deliver more than 150 mg of toxins in a headbutt (which is a lot—on par with how much venom a king cobra delivers in a single bite!), there’s another wrinkle with regards to deciding what toxins are the most potent: humans aren’t mice.

Almost every LD50 is calculated in mice or rats, as they’re easy to generate and care for and are small enough (by weight) that you don’t need a large amount of toxin to conduct lethality experiments. That said, there are plenty of exceptions; LD50 experiments have been run on everything from cockroaches to cats. And the fact is, you simply can’t assume that a dose that is lethal to one species will be equally lethal to another species, or perhaps most importantly, to our species. Different species process toxins differently, which is something important for pet owners to keep in mind they stock their home with Tylenol.

Great for pregnant women, deadly for cats. Photo by Austin Kirk

Great for pregnant women, deadly for cats. Photo by Austin Kirk

Acetaminophen, the active ingredient in the painkiller Tylenol, is considered one of the safest drugs for humans to take. It’s considered so safe that it’s the go-to painkiller for pregnant women. But in dogs and cats, even small amounts can be deadly. The instructions on my large bottle of extra-strength, rapid release acetaminophen say that I, as an adult human, can take two 500 mg pills every six hours, which is a dose of approximately 12.5 mg/kg. Acute toxicity isn’t likely unless I ingest about 15 pills, receiving a dose of 93.75 mg/kg (you can assume lethal doses are even higher than that). In dogs and cats, however, acute toxicity occurs at much lower amounts; 75 mg/kg in dogs and only 10 mg/kg in cats, much less than the recommended dose for people. Lethal doses can occur at as low as 500 mg/kg in dogs, and 50 mg/kg in cats. Meanwhile, the LD50 in rats, rabbits and guinea pigs is over 2000 mg/kg.

Species-specific effects are seen in all sorts of toxins, so if you’re asking what the deadliest venom or poison is, it matters what animals were used to conduct the LD50 studies. We can compare studies done in the same creatures, but that doesn’t tell us what toxin is deadliest in the species we care most about: us. Since we don’t conduct LD50 tests in humans (for obvious ethical reasons), we actually don’t know what the deadliest toxin is to humans, intravenously, subcutaneously, or any-ously. (We can make educated guesses, but ultimately, we’re not 100% sure, which is why agencies like the FDA and EPA build expansive error margins into their calculations of how much of a chemical is “safe”—usually, at least ten times less than the lowest dose seen to cause any kind of damage in animals.)

But I would argue that even if we judge “deadliest” by a particular route of administration into a particular species of animal, the title is still meaningless. Just because those frogs possess a toxin that can kill mice in the lab doesn’t mean we should freak out if we’re locked in a room with one, or even grabbing one bare-handed. By my estimations of how much toxin each animal has, it would take hundreds of amphibian head butts to even come close to delivering a lethal dose to a person—probably more like thousands. There are plenty of species much more likely to kill you if you do end up on the wrong end of their toxic weaponry.

Gambling with your life

Perhaps a better measure of deadliness is what is referred to as mortality or fatality rate: the likelihood of death following toxin exposure. Or, in plainer terms: if the creature in front of me bites or stings me, what are my odds?

It’s clear by this measure, the frogs I discussed in the last section are harmless. Other venomous animals are decidedly not. As a group, venomous snakes are the ones people probably fear the most, and some of that fear is for good reason: untreated, many species’ bites carry fatality rates between 60 and 100%. However, since the advent of antivenom, those rates have dropped dramatically in many areas. In Australia for instance, although the country is notorious for having some of the ‘deadliest’ snakes in the world (by LD50 measures), only about 0.001% of the 1,000+ people who are envenomated by snakes each year die thanks to effective and expedient medical care. Of course, there are parts of the world where such resources are much harder to come by, and medical politics, cultural practices, and lack of rapid access to affordable care make snakebites much more dangerous. And some species still have high odds of killing you, either because of the nature of their venoms or the amount they inject. So whether you should worry about a snakebite really depends on where you are and what resources you have readily available.

Map showing the number of global snake envenomings, based on data by Kasturiratne et al. 2008.

Map showing the number of global snake envenomings—only some of which are fatal—based on data by Kasturiratne et al. 2008.

Snakes aside, there are other animals whose bites and stings are life-threatening emergencies. The geographer’s cone (Conus geographus) has an estimated fatality rate of 70%, which reflects the speed of its paralytic venom: death by snail sting tends to occur in a matter of minutes. Blue ringed octopus bites (Hapalochlaena spp.) can also lead to incurable, life-threatening paralysis thanks to the presence of tetrodotoxin in their venom. But most venomous species, whether you’re talking sea creatures, spiders, scorpions, or snakes, actually have fairly low fatality rates—often below 1% (albeit a 1% chance of death is still high enough to be worth seeking immediate medical attention!).

If you really want to talk about highest fatality rates from toxin exposure, many poisonous species make their venomous counterparts look downright benign. Down a pufferfish without removing its organs (tetrodotoxin), the wrong kind of “magic” mushroom (amatoxin), or simply improperly sterilized canned foods (Botulinum toxin), and your odds of survival could be slim to none. That said, since exposure to poisons generally requires some effort on the victim’s part (or on the part of a would be assassin), even the deadliest poisonous creatures on Earth rarely inspire the same kind of fear as less-deadly venomous ones.

A winning straight-up wager pays 35-1, but you're more likely to score on a column or dozens bet. Image credit: Wikimedia Commons

A winning straight up wager pays 35-1, but you’re more likely to score on a column or dozens bet. Image credit: Wikimedia Commons

But there is a one glaring problem with measuring deadliness by fatality rate: while the odds of dying once exposed to the toxin may be high, the odds of encountering the toxins of many deadly species is vanishingly small. The odds that the average human being will spend any amount of time in their life within ten feet of a king cobra or a geographer’s cone snail, let alone experience the venom of one, are slim. So calling these species the “deadliest” when compared to other toxic species is kind of like saying to win money, you should always place a straight up wager rather than a dozens or column bet in roulette. Sure, with a straight up wager you’ll win big if the ball lands on your number, but the odds of that event happening are much lower than the odds of the ball landing on one of the 12 numbers in a dozens or column bet. The multi-number bets pay less per hit, but those hits are much more likely to happen.

A number’s game

Statistically speaking, the title of deadliest toxic animal on the planet should go to the one that causes the most deaths every year because your odds of dying by that animal are numerically the greatest. If animal A kills ten people worldwide every year, and animal B kills 10 million, then the average person is way more likely to be killed by animal B. Period.

We can break such numbers down by country or region; after all, you have nothing to fear from an animal that doesn’t live where you do. If we do that, then places like the U.S., Australia, and Europe have essentially nothing to fear from snakes, as they kill only a handful of people in each annually. Far deadlier are bees, ants, and wasps—members of the insect order Hymenoptera—whose venoms kill dozens of people in each of those regions every year due to allergic reactions. But there are even deadlier killers, found on every continent except Antarctica, whose toxic bites cause more deaths annually than any other animals: mosquitoes.

A female Culex quinquefasciatus, one of the species of mosquito that spreads West Nile Virus. Photo credit: CDC/ James Gathany

A female Culex quinquefasciatus, one of the species of mosquito that spreads West Nile Virus. Photo credit: CDC/ James Gathany

If we talk numbers, then mosquito bites—thanks to the deadly pathogens they can vector—kill somewhere between 750,000 and 1,000,000 people every year (and yes, mosquitoes are venomous—and their venom proteins may even facilitate disease transmission!). Even in the U.S., where the biggest vector-borne threats like malaria and yellow fever are effectively eradicated, dozens to hundreds of deaths occur from other mosquito-delivered diseases like West Nile Virus, putting the death toll from mosquitoes ahead of spiders, snakes, scorpions, bees and wasps combined. And in other parts of the world, even ones where snakebites take a devastating toll, mosquito bites out-kill every other toxic group. In Africa, for example, snakebites may take a staggering 30,000 to 40,000 lives every year. But that number pales in comparison to the approximately 400,000 people who die there as the result of mosquito-borne malaria—and malaria is only one of the possible mosquito-carried diseases that Africans have to worry about.

So is it the deadliest, deadliest or deadliest?

There you have it: the three main ways to assess deadliness, and the toxic species or groups that dominate them. Toxic species can be deadly by the dose, deadly if exposed, and/or deadly in our daily lives. So the next time you’re wondering what is the deadliest plant or animal out there, ask yourself what you really want to know. Academically determining what toxins have potent effects on our bodies is very different from assessing your personal risk. And take comfort in the fact that, overall, you can protect yourself against the toxic creatures most likely to kill you every day with one simple thing: insect repellent.


  • Uncle Al

    Acetaminophen is substantially hepatotoxic for being rapidly metabolized to N-acetylquinonimine (brick red color – crushed Tylenol in vinegar, add a sprinkle of sodium nitrite, swirl). It is then immediately conjugated with glutathione to be excreted,

    Substantial hepatic glutathione depletion occurs through binge drinking and fasting, both of which cause headaches. One also wonders about Acetoaminophen plus nitrite-cured meat in gastric juice.

  • Kaleberg

    Thanks for putting things into perspective. There’s a tendency to overrate intensity over omnipresence, possibly because one can at least try and avoid particularly dangerous things, but you often have no choice but to travel by car.

  • Marcel Volker

    The map showing number of envenomings is somewhat misleading. The original shows the number by region (e.g. Europe), but your map puts country borders back in.
    So now it appears that every country in Europe has over a thousand snake envenomings a year – which is wildly unlikely on the face of it (given countries like the UK and the Netherlands IIRC only have a single, rare, shy, venomous species, the adder). Also, the Nordic countries at over 1000 each, with their climate, is implausible.


Science Sushi

Real Science. Served Raw.

About Christie Wilcox

Dr. Christie Wilcox is a science writer and postdoctoral scholar at the University of Hawaii. She freelances for major media outlets including The New York Times and Popular Science. Her debut book, Venomous, releases August 2016 (Scientific American/FSG Books). To learn more about her life and work, check out her webpage or follow her on Twitter, Google+, or Facebook.


See More


@NerdyChristie on Twitter

Discover's Newsletter

Sign up to get the latest science news delivered weekly right to your inbox!

Collapse bottom bar