'The Worlds Most Venomous Spider'
I thought people might like a read into this subjects complexities and why there is often so much confusion, including some of the traps people fall into when debating the topic.
I personally have been researching and studying medically significant spiders for over 20 years. I would like to thank Dr Volker Herzig and Richard Vetter for their review and input into this also.
(This was originally a video transcript so sources aren't provided in the text, however if anyone would like any of the source papers let me know and I'll be happy to provide them.)
So,
'What is the world's most venomous spider?'
It's a question I'm asked often, and I see many videos or comments online all confidently giving answers that contradict each other.
The problem is that people want a simple answer to a complex question.
I'm going to explain why there appears to be so much confusion on the subject.
People also ask, "What spider is the most dangerous?", which is slightly different and takes into consideration things like the temperament of spiders, their likelihood of biting, and their proximity to people. But, we'll keep it simple and look at the relative dangers of each spider if they were to bite you.
So, "What is the most venomous spider?"
The first issue we encounter is the question "What is the most venomous spider?" doesn’t have a set meaning.
We could use some literal dictionary definitions of 'venomous', i.e., "capable of injecting venom by means of a bite or sting."
Then the question posed is, "What spider is the most capable of injecting venom?".
There are thousands of spiders equally capable of injecting venom, but I know that isn't what people are asking.
What people usually mean is either:
1. "Which spider has the most potent venom?" or
2. "Which spider has the most harmful bite?"
These questions are subtly different. The first looks at which animal's venom is the most potent drop for drop. The second not only looks at the venom's potency, but also considers the amount. Some animals have a weaker venom, but inject more of it, making bites more harmful as a whole.
If we were to compare using the latter, what amount do we use? Do we take the average amount an adult typically injects per wet bite? Do we use the maximum an adult could hypothetically inject, i.e., its maximum yield (even though they never realistically use this much)? Or, If we want to work out an average across all bites, then we'll need to include dry bites (bites that contain no venom). Dry bites make up a significant portion of the defensive bites from some species more so than others, and including these will significantly alter your figures.
You also face further complications with this method, such as varying toxicity levels and yields between the sexes.
So, maybe venom potency drop for drop is the best way?
This brings us to LD50s.
An LD50 or 'lethal dose 50', is essentially the amount of venom required to kill 50% of the animals in a test group. For example, you have 100 mice, and inject each of them with 1 mg of venom. If at that dose 50 of the mice die while 50 live, 1mg is your LD50.
If at 1mg more than 50 die, you would lower the dose and test again. If less than 50 die, you would increase the dose and so on. This process continues until you find the amount that kills exactly 50%.
(*In reality labs often won't reach exactly 50%. They may calculate an LD50 based on results at higher or lower doses*)
The reason we use 50% is that some animals in a group will naturally have a higher tolerance, and some will naturally have a lower tolerance to the venom.
At the amount where 50% die, this is obviously a fair representation of the average amount required to kill a member of that species.
The lower the dose, the more potent the venom.
To make this figure easily comparable against other results often involving different test subjects, we round it up or down and express it in milligrams per kilogram of bodyweight. If 1mg was the dose where 50% of our mice died, and each mouse weighed 20 grams; with 1,000 grams in a kilogram, we multiply that 1mg by 50 to give us a standardised LD50 of 50mg per kg.
So far, so good...
However, there is no single LD50 for each target species the venom is tested on. This figure varies greatly depending on the route of administration. There are numerous ways venom can be administered: subcutaneously (below the epidermis), intravenously (directly into a vein), intramuscularly (into muscle tissue), intradermally (just below the dermis or upper layers of skin), intraperitoneally (into the peritoneal cavity where your organs are), and intracerebroventricularly (into the cerebral ventricles, essentially directly into the brain, bypassing the blood-brain barrier).
This is something many online videos or articles fail to recognise. I often see the subcutaneous LD50 results from one species being compared with the intracerebroventricular LD50 of another. Comparing the lethality rate of two neurotoxic substances where one has been administered under the skin and the other has been injected directly into the brain, isn’t going to give you an accurate representation of their relative potency.
This is like comparing the lethality of bullets by shooting 10 people in the head with a 9mm, 10 in the foot with a .50 caliber machine gun, then declaring that 9mm bullets are more dangerous because more people died when they were shot with them.
Aside from the differing methods of administration, there is also a huge difference in the results depending on which animal is being used as a test subject. Spider venoms are extremely complex substances comprised of hundreds of individual components which all act on different chemical processes in a victim’s body. Different animals have different physiological processes. Some parts of a spider’s venom might significantly disrupt the processes in one animal and not the next. Venoms affect different animals in different ways, and the dose required to kill individual species varies. There is no universal "one size fits all." Asking "What venom is the most toxic?" is a bit like asking a doctor "Which medicine is the best?".
Without setting any criteria, it's meaningless.
I've always assumed that when people ask me, 'What is the most venomous spider?' they are referring to its effects on humans. Determining the most harmful or toxic spider towards one animal is challenging enough. Anyone who claims to know the individual answer for every mammal, bird, fish, reptile, amphibian and arthropod on Earth, and has calculated a definitive answer overall, is lying.
Most LD50 studies use adult mice. Mice are mammals like us and have many similar biological processes. However, we are not mice. While there are similarities, our physiology is ultimately different from that of a mouse. This animal is really an arbitrary creature we frequently test venoms on because aside from their similarities to us, they’re small, easy to breed, and relatively easy to care for in a lab. Yes, mice can give an indication of the potential harms of a substance, but they cannot be used as a direct like-for-like replacement to show exactly how something will affect us.
An excellent example of this is to compare the LD50 of two spiders often towards the top of many "most venomous" lists: the Sydney funnel-web (Atrax robustus) and the Brazilian wandering spider (Phoneutria nigriventer).
The subcutaneous LD50 of Phoneutria venom when tested on adult mice is consistently shown to be around 0.7 mg/kg. That’s 0.7 mg of venom needed to kill a kilogram of mice.
This same test from a study using Atrax venom gave a result of around 11.3 mg/kg for the male spider’s venom and 80 mg/kg for the female’s. If we use mice as the benchmark, then Brazilian wandering spider venom would be considered between 16 and 114 times more toxic than a Sydney funnel-webs.
We know this isn’t an accurate reflection of their toxicity towards us because:
1. They wouldn’t pose a danger to us if this figure translated to humans, and
2. Funnel-web LD50 studies have been done on macaques (which are primates much more closely related to us than mice are), and the results were drastically different.
The subcutaneous LD50 of male funnel-web venom when tested on these animals was below 0.2 mg/kg. Sydney funnel-web venom is over 56 times more harmful to a monkey than it is to a mouse.
It's important to note there have never been any LD50 studies done on primates using wandering spider venom, so we are unable to directly compare the two.
Sometimes, even when the comparisons appear fair, i.e., both done on mice, both using the same method, there are still incomparable figures published side by side.
0.16 mg/kg is often used in tables as the subcutaneous LD50 for Sydney funnel-web venom in mice.
However, when you read the source material, you see this was the LD50 for a single isolated component of funnel-web venom, Robustoxin, not the whole venom, and only when tested on mice less than 2 days old.
While not a direct equivalence; to illustrate this, imagine comparing the strength of different alcoholic drinks: wine, vodka, whisky, etc.
You find that beer is the strongest, but fail to mention that unlike the other drinks, only the ethanol in beer was used while conducting your study. Not only this, but while all other drinks were given to adults, your pure 'beer ethanol' was given to a baby.
The same study found the crude, or 'whole' venom was 10 times less effective on these same baby mice, with an LD50 of 1.5 mg/kg. This figure then dramatically increased as soon as they left this early stage in their life.
Even with total consistency across all tests, trying to obtain a single figure to use in comparisons is tricky. Aside from the natural fluctuation between individual specimens, studies show that venom composition and potency varies within a species depending on the region. Considering the large range of some of these spiders, results will differ significantly depending on where your specimens were collected.
This is also true for test subjects, with different populations of the same species varying in their resistance to venoms. One supplier sells 46 different strains of mice each with differing results.
In the late 20th century two experiments were completed by two separate labs on adult mice. Both used the same method, and both used the same species of wandering spider venom. One lab found an intravenous LD50 of 0.57mg/kg, while the other had a result of 0.34mg/kg. That's a 70% increase in the venoms apparent potency, for what should on paper be the same result.
Different populations of both spider and test subject, along with tiny differences in how the same test was carried out, had a significant impact on the result.
As a side note, due to the increasing concern around the ethics of killing large numbers of animals in a lab, a lot of these studies are now quite old. This can cause complications when outdated taxonomic names are published alongside results. For example, with Brazilian Wandering spiders, I often see 'Phoneutria fera' linked with mice LD50 results.
In reality, the spiders involved in these tests were either Phoneutria nigriventer or Phoneutria keyserlingi.
At the time, these three spiders were not taxonomically separated (recognised as separate species). As 'fera' is the type species for the genus, the other two were lumped together under this name.
Since they have been separated, it's clear fera could never have been involved.
The specimens used in these studies were collected around the Butantan institute in São Paulo, where only nigriventer and keyserlingi are found. feras range starts about a thousand miles north.
There has never been any mammalian LD50 studies done using Phoneutria fera venom.
Since these revisions, the authors have corrected this, however it appears to have gone largely unnoticed.
LD50 results also sometimes conflict with other data we can use to try and determine the most harmful bite to humans, even within members of the same genus. The highest mortality rate from South American recluse spider bites is reported in an area where Chilean recluse spiders (Loxosceleslaeta) are most prevalent, but when looking at mice LD50 results, this species had a significantly less effective venom than its sister species, the Brazilian brown recluse (Loxoscelesintermedia).
The intraperitoneal LD50 for mice using Chilean recluse venom was 1.45 mg/kg, while Brazilian brown recluse venom was three times more effective at 0.48 mg/kg.
This leads on to the next factor we need to consider when looking at the potential harms of a spider bite: where people are bitten.
Some venoms are much more adept at making their way through tissue than others. Recluse spider venom is an excellent example of this. Due to its cytotoxic nature, the venom is efficient at breaking down cells and making its way through the body. A pure neurotoxin, no matter how potent, would be far less efficient if injected into a fatty part of the body than if injected directly into or close to the bloodstream. It would be absorbed too slowly, giving the body an opportunity to break it down before it can cause harm. Spider venoms vary in this capacity, and this is reflected in LD50 results. Some have a more consistent figure when compared across a range of methods, while others show more significant differences depending on the route of administration. The same dose of venom from the same bite may be far more dangerous in one area than it would another.
If we want to compare, where do we choose as a standard bite location? The hand? The thigh? The soft tissue on the inside of the wrist?
Different areas may significantly hinder or benefit the efficiency of some venoms and not others.
There is another way of looking at the potential lethality of spider bites towards humans: the number of bites recorded against the number of deaths.
This data has its own complications.
Brazilian wandering spiders have a proven low LD50 when tested across a broad range of animals. They have a potent venom, a high venom yield, and the capacity to give a relatively deep bite. South American recluse spiders on the other hand, while they do have a relatively low LD50, also have small fangs and a low venom yield.
These spiders only have the capacity to store around 0.06 mg of venom (compared to wandering spiders, which on average yield about 20 times that amount). Yet, recent figures show South American recluse spiders cause more deaths per bite than wandering spiders do.
So why is this? Do they inject venom in more bites? Would their LD50, if tested on humans, be lower? Is their venom more efficient at spreading from the areas people are typically bitten?
These may all be legitimate reasons, but there are other, more human factors to consider.
Firstly, a victim's inclination to seek early medical treatment. Wandering spiders are large spiders that give a painful bite. You know if you've been bitten by one, and due to the immediate pain, victims may be more inclined to attend hospital sooner. Loxosceles bites on the other hand make far less of an initial impact, often going ignored, or with victims not recognising symptoms until it's too late.
Secondly, due to the nature of this spider's toxin, the current antivenom isn't as effective as it is for other spiders. Even with medical treatment, there remains a higher likelihood of adverse outcomes. Having little data on the death rate for many significant species prior to the development of antivenom makes comparing what their mortality rate would be without it quite challenging. Antivenom is given as a precaution to help people who start displaying more serious symptoms, but this doesn't necessarily mean they would have died without it.
Thirdly, data is only as good as what's recorded. This brings me onto a recent poll I conducted.
A lot of these studies rely on people attending hospitals to have their bite recorded. Some people go as a precaution as soon as a bite occurs, but others won't attend unless they develop symptoms and feel they need it.
I asked people this question using a black widow bite as an example. 70% said they would go straight away, while 30% said they would only go if they felt it was necessary.
This poll was far from a scientific study, but it gives an indication that a large number of asymptomatic or mild bites likely go unreported. This percentage will vary depending on the type of spider and area the study was conducted, disproportionately effecting statistics. Misidentification is also common in these data sets, which usually rely on medical staff or the public to correctly identify spiders.
On top of this, some studies try and concentrate on the results from a specific species, while others group a whole genus together, including the potentially less harmful members.
The animals we often want to compare come from different parts of the world, with studies being done by different institutions in different continents. There's no global organization governing results to ensure they're counted consistently.
Maybe we can look at the number of serious bites or bites that required antivenom as an indicator? We encounter similar issues here. Different studies include different groups of people, and different institutions grade bite severity differently.
Some articles quote severe symptoms from black widow envenomation as high as 54%, but when you read the source, that 54% was within a group of people that had been referred to a specialist toxicology unit. Of course there's going to be a disproportionately high number of patients with severe symptoms in there; that's why they're in there.
This is like stating 80% of people have life-threatening symptoms after contracting COVID, but failing to mention that figure was from a study of patients in the intensive care unit on a COVID ward.
On the contrary, another study showed severe reactions to black widow bites in the US as low as 0.5%. This figure, taken from poison control centers, includes a large number of 'suspected exposures'. This includes people who phoned in believing they had been bitten. I've seen the identification skills of most people, and at best it's poor. Aswell as cases of misidentification, add to this the number of people who developed a rash or pimple and diagnosed themselves as being bitten by a black widow, and I can't imagine this figure is entirely accurate either.
If we look at this same study and only count those who attended a medical facility, then of those 830, 13 had major effects. Instead of 0.5%, this now gives us a severe envenomation rate of 1.5%.
As previously noted, not everyone who was actually bitten will have attended, so the true percentage of severe envenomation is likely to sit somewhere between the two.
continues below....
I thought people might like a read into this subjects complexities and why there is often so much confusion, including some of the traps people fall into when debating the topic.
I personally have been researching and studying medically significant spiders for over 20 years. I would like to thank Dr Volker Herzig and Richard Vetter for their review and input into this also.
(This was originally a video transcript so sources aren't provided in the text, however if anyone would like any of the source papers let me know and I'll be happy to provide them.)
So,
'What is the world's most venomous spider?'
It's a question I'm asked often, and I see many videos or comments online all confidently giving answers that contradict each other.
The problem is that people want a simple answer to a complex question.
I'm going to explain why there appears to be so much confusion on the subject.
People also ask, "What spider is the most dangerous?", which is slightly different and takes into consideration things like the temperament of spiders, their likelihood of biting, and their proximity to people. But, we'll keep it simple and look at the relative dangers of each spider if they were to bite you.
So, "What is the most venomous spider?"
The first issue we encounter is the question "What is the most venomous spider?" doesn’t have a set meaning.
We could use some literal dictionary definitions of 'venomous', i.e., "capable of injecting venom by means of a bite or sting."
Then the question posed is, "What spider is the most capable of injecting venom?".
There are thousands of spiders equally capable of injecting venom, but I know that isn't what people are asking.
What people usually mean is either:
1. "Which spider has the most potent venom?" or
2. "Which spider has the most harmful bite?"
These questions are subtly different. The first looks at which animal's venom is the most potent drop for drop. The second not only looks at the venom's potency, but also considers the amount. Some animals have a weaker venom, but inject more of it, making bites more harmful as a whole.
If we were to compare using the latter, what amount do we use? Do we take the average amount an adult typically injects per wet bite? Do we use the maximum an adult could hypothetically inject, i.e., its maximum yield (even though they never realistically use this much)? Or, If we want to work out an average across all bites, then we'll need to include dry bites (bites that contain no venom). Dry bites make up a significant portion of the defensive bites from some species more so than others, and including these will significantly alter your figures.
You also face further complications with this method, such as varying toxicity levels and yields between the sexes.
So, maybe venom potency drop for drop is the best way?
This brings us to LD50s.
An LD50 or 'lethal dose 50', is essentially the amount of venom required to kill 50% of the animals in a test group. For example, you have 100 mice, and inject each of them with 1 mg of venom. If at that dose 50 of the mice die while 50 live, 1mg is your LD50.
If at 1mg more than 50 die, you would lower the dose and test again. If less than 50 die, you would increase the dose and so on. This process continues until you find the amount that kills exactly 50%.
(*In reality labs often won't reach exactly 50%. They may calculate an LD50 based on results at higher or lower doses*)
The reason we use 50% is that some animals in a group will naturally have a higher tolerance, and some will naturally have a lower tolerance to the venom.
At the amount where 50% die, this is obviously a fair representation of the average amount required to kill a member of that species.
The lower the dose, the more potent the venom.
To make this figure easily comparable against other results often involving different test subjects, we round it up or down and express it in milligrams per kilogram of bodyweight. If 1mg was the dose where 50% of our mice died, and each mouse weighed 20 grams; with 1,000 grams in a kilogram, we multiply that 1mg by 50 to give us a standardised LD50 of 50mg per kg.
So far, so good...
However, there is no single LD50 for each target species the venom is tested on. This figure varies greatly depending on the route of administration. There are numerous ways venom can be administered: subcutaneously (below the epidermis), intravenously (directly into a vein), intramuscularly (into muscle tissue), intradermally (just below the dermis or upper layers of skin), intraperitoneally (into the peritoneal cavity where your organs are), and intracerebroventricularly (into the cerebral ventricles, essentially directly into the brain, bypassing the blood-brain barrier).
This is something many online videos or articles fail to recognise. I often see the subcutaneous LD50 results from one species being compared with the intracerebroventricular LD50 of another. Comparing the lethality rate of two neurotoxic substances where one has been administered under the skin and the other has been injected directly into the brain, isn’t going to give you an accurate representation of their relative potency.
This is like comparing the lethality of bullets by shooting 10 people in the head with a 9mm, 10 in the foot with a .50 caliber machine gun, then declaring that 9mm bullets are more dangerous because more people died when they were shot with them.
Aside from the differing methods of administration, there is also a huge difference in the results depending on which animal is being used as a test subject. Spider venoms are extremely complex substances comprised of hundreds of individual components which all act on different chemical processes in a victim’s body. Different animals have different physiological processes. Some parts of a spider’s venom might significantly disrupt the processes in one animal and not the next. Venoms affect different animals in different ways, and the dose required to kill individual species varies. There is no universal "one size fits all." Asking "What venom is the most toxic?" is a bit like asking a doctor "Which medicine is the best?".
Without setting any criteria, it's meaningless.
I've always assumed that when people ask me, 'What is the most venomous spider?' they are referring to its effects on humans. Determining the most harmful or toxic spider towards one animal is challenging enough. Anyone who claims to know the individual answer for every mammal, bird, fish, reptile, amphibian and arthropod on Earth, and has calculated a definitive answer overall, is lying.
Most LD50 studies use adult mice. Mice are mammals like us and have many similar biological processes. However, we are not mice. While there are similarities, our physiology is ultimately different from that of a mouse. This animal is really an arbitrary creature we frequently test venoms on because aside from their similarities to us, they’re small, easy to breed, and relatively easy to care for in a lab. Yes, mice can give an indication of the potential harms of a substance, but they cannot be used as a direct like-for-like replacement to show exactly how something will affect us.
An excellent example of this is to compare the LD50 of two spiders often towards the top of many "most venomous" lists: the Sydney funnel-web (Atrax robustus) and the Brazilian wandering spider (Phoneutria nigriventer).
The subcutaneous LD50 of Phoneutria venom when tested on adult mice is consistently shown to be around 0.7 mg/kg. That’s 0.7 mg of venom needed to kill a kilogram of mice.
This same test from a study using Atrax venom gave a result of around 11.3 mg/kg for the male spider’s venom and 80 mg/kg for the female’s. If we use mice as the benchmark, then Brazilian wandering spider venom would be considered between 16 and 114 times more toxic than a Sydney funnel-webs.
We know this isn’t an accurate reflection of their toxicity towards us because:
1. They wouldn’t pose a danger to us if this figure translated to humans, and
2. Funnel-web LD50 studies have been done on macaques (which are primates much more closely related to us than mice are), and the results were drastically different.
The subcutaneous LD50 of male funnel-web venom when tested on these animals was below 0.2 mg/kg. Sydney funnel-web venom is over 56 times more harmful to a monkey than it is to a mouse.
It's important to note there have never been any LD50 studies done on primates using wandering spider venom, so we are unable to directly compare the two.
Sometimes, even when the comparisons appear fair, i.e., both done on mice, both using the same method, there are still incomparable figures published side by side.
0.16 mg/kg is often used in tables as the subcutaneous LD50 for Sydney funnel-web venom in mice.
However, when you read the source material, you see this was the LD50 for a single isolated component of funnel-web venom, Robustoxin, not the whole venom, and only when tested on mice less than 2 days old.
While not a direct equivalence; to illustrate this, imagine comparing the strength of different alcoholic drinks: wine, vodka, whisky, etc.
You find that beer is the strongest, but fail to mention that unlike the other drinks, only the ethanol in beer was used while conducting your study. Not only this, but while all other drinks were given to adults, your pure 'beer ethanol' was given to a baby.
The same study found the crude, or 'whole' venom was 10 times less effective on these same baby mice, with an LD50 of 1.5 mg/kg. This figure then dramatically increased as soon as they left this early stage in their life.
Even with total consistency across all tests, trying to obtain a single figure to use in comparisons is tricky. Aside from the natural fluctuation between individual specimens, studies show that venom composition and potency varies within a species depending on the region. Considering the large range of some of these spiders, results will differ significantly depending on where your specimens were collected.
This is also true for test subjects, with different populations of the same species varying in their resistance to venoms. One supplier sells 46 different strains of mice each with differing results.
In the late 20th century two experiments were completed by two separate labs on adult mice. Both used the same method, and both used the same species of wandering spider venom. One lab found an intravenous LD50 of 0.57mg/kg, while the other had a result of 0.34mg/kg. That's a 70% increase in the venoms apparent potency, for what should on paper be the same result.
Different populations of both spider and test subject, along with tiny differences in how the same test was carried out, had a significant impact on the result.
As a side note, due to the increasing concern around the ethics of killing large numbers of animals in a lab, a lot of these studies are now quite old. This can cause complications when outdated taxonomic names are published alongside results. For example, with Brazilian Wandering spiders, I often see 'Phoneutria fera' linked with mice LD50 results.
In reality, the spiders involved in these tests were either Phoneutria nigriventer or Phoneutria keyserlingi.
At the time, these three spiders were not taxonomically separated (recognised as separate species). As 'fera' is the type species for the genus, the other two were lumped together under this name.
Since they have been separated, it's clear fera could never have been involved.
The specimens used in these studies were collected around the Butantan institute in São Paulo, where only nigriventer and keyserlingi are found. feras range starts about a thousand miles north.
There has never been any mammalian LD50 studies done using Phoneutria fera venom.
Since these revisions, the authors have corrected this, however it appears to have gone largely unnoticed.
LD50 results also sometimes conflict with other data we can use to try and determine the most harmful bite to humans, even within members of the same genus. The highest mortality rate from South American recluse spider bites is reported in an area where Chilean recluse spiders (Loxosceleslaeta) are most prevalent, but when looking at mice LD50 results, this species had a significantly less effective venom than its sister species, the Brazilian brown recluse (Loxoscelesintermedia).
The intraperitoneal LD50 for mice using Chilean recluse venom was 1.45 mg/kg, while Brazilian brown recluse venom was three times more effective at 0.48 mg/kg.
This leads on to the next factor we need to consider when looking at the potential harms of a spider bite: where people are bitten.
Some venoms are much more adept at making their way through tissue than others. Recluse spider venom is an excellent example of this. Due to its cytotoxic nature, the venom is efficient at breaking down cells and making its way through the body. A pure neurotoxin, no matter how potent, would be far less efficient if injected into a fatty part of the body than if injected directly into or close to the bloodstream. It would be absorbed too slowly, giving the body an opportunity to break it down before it can cause harm. Spider venoms vary in this capacity, and this is reflected in LD50 results. Some have a more consistent figure when compared across a range of methods, while others show more significant differences depending on the route of administration. The same dose of venom from the same bite may be far more dangerous in one area than it would another.
If we want to compare, where do we choose as a standard bite location? The hand? The thigh? The soft tissue on the inside of the wrist?
Different areas may significantly hinder or benefit the efficiency of some venoms and not others.
There is another way of looking at the potential lethality of spider bites towards humans: the number of bites recorded against the number of deaths.
This data has its own complications.
Brazilian wandering spiders have a proven low LD50 when tested across a broad range of animals. They have a potent venom, a high venom yield, and the capacity to give a relatively deep bite. South American recluse spiders on the other hand, while they do have a relatively low LD50, also have small fangs and a low venom yield.
These spiders only have the capacity to store around 0.06 mg of venom (compared to wandering spiders, which on average yield about 20 times that amount). Yet, recent figures show South American recluse spiders cause more deaths per bite than wandering spiders do.
So why is this? Do they inject venom in more bites? Would their LD50, if tested on humans, be lower? Is their venom more efficient at spreading from the areas people are typically bitten?
These may all be legitimate reasons, but there are other, more human factors to consider.
Firstly, a victim's inclination to seek early medical treatment. Wandering spiders are large spiders that give a painful bite. You know if you've been bitten by one, and due to the immediate pain, victims may be more inclined to attend hospital sooner. Loxosceles bites on the other hand make far less of an initial impact, often going ignored, or with victims not recognising symptoms until it's too late.
Secondly, due to the nature of this spider's toxin, the current antivenom isn't as effective as it is for other spiders. Even with medical treatment, there remains a higher likelihood of adverse outcomes. Having little data on the death rate for many significant species prior to the development of antivenom makes comparing what their mortality rate would be without it quite challenging. Antivenom is given as a precaution to help people who start displaying more serious symptoms, but this doesn't necessarily mean they would have died without it.
Thirdly, data is only as good as what's recorded. This brings me onto a recent poll I conducted.
A lot of these studies rely on people attending hospitals to have their bite recorded. Some people go as a precaution as soon as a bite occurs, but others won't attend unless they develop symptoms and feel they need it.
I asked people this question using a black widow bite as an example. 70% said they would go straight away, while 30% said they would only go if they felt it was necessary.
This poll was far from a scientific study, but it gives an indication that a large number of asymptomatic or mild bites likely go unreported. This percentage will vary depending on the type of spider and area the study was conducted, disproportionately effecting statistics. Misidentification is also common in these data sets, which usually rely on medical staff or the public to correctly identify spiders.
On top of this, some studies try and concentrate on the results from a specific species, while others group a whole genus together, including the potentially less harmful members.
The animals we often want to compare come from different parts of the world, with studies being done by different institutions in different continents. There's no global organization governing results to ensure they're counted consistently.
Maybe we can look at the number of serious bites or bites that required antivenom as an indicator? We encounter similar issues here. Different studies include different groups of people, and different institutions grade bite severity differently.
Some articles quote severe symptoms from black widow envenomation as high as 54%, but when you read the source, that 54% was within a group of people that had been referred to a specialist toxicology unit. Of course there's going to be a disproportionately high number of patients with severe symptoms in there; that's why they're in there.
This is like stating 80% of people have life-threatening symptoms after contracting COVID, but failing to mention that figure was from a study of patients in the intensive care unit on a COVID ward.
On the contrary, another study showed severe reactions to black widow bites in the US as low as 0.5%. This figure, taken from poison control centers, includes a large number of 'suspected exposures'. This includes people who phoned in believing they had been bitten. I've seen the identification skills of most people, and at best it's poor. Aswell as cases of misidentification, add to this the number of people who developed a rash or pimple and diagnosed themselves as being bitten by a black widow, and I can't imagine this figure is entirely accurate either.
If we look at this same study and only count those who attended a medical facility, then of those 830, 13 had major effects. Instead of 0.5%, this now gives us a severe envenomation rate of 1.5%.
As previously noted, not everyone who was actually bitten will have attended, so the true percentage of severe envenomation is likely to sit somewhere between the two.
continues below....
.
