Spider Tales 7. the Hobo Spider: Anatomy of a Miscarriage of Justice

"What they want, you see, is not the truth; it's coherence. A miscarriage of justice is always a masterpiece of coherence."
Daniel Pennac, Monsieur Malaussène (1995)*

 

Among the common spiders in Europe, giant house spiders (genus Eratigena, particularly species of the Eratigena atrica group) certainly are the most dreaded by arachnophobes. Huge, long-legged and dazzlingly fast, male giant house spiders often cause quite a commotion when they bolt across rooms at dusk in summer and fall, as they wander through homes in search of a mate. However, the fear and disgust they tend to induce in many people is purely based on their dark, leggy and hairy appearance, and their large size; pretty much everyone knows these impressive spiders are quite harmless. Even the most depraved tabloids, despite their tendency to make stories out of the silliest rumours, do not dare to claim they are dangerous. Their bites are not, and have never been, a cause for concern.

Reaching up to 12.5 cm (almost 5 inches) legs included, male giant house spiders (Eratigena gr.atrica) are among the largest European spiders in terms of legspan

 

The situation in North America, on the other hand, is the complete opposite: an established, non-native Eratigena from Europe used to be widely feared and blamed for necrotic bites in the Western states.
For a few decades, it was included, alongside the other eight-legged desperados of the continent, the recluses and widows, in the list of medically significant spiders in the USA... Until it was removed from that list ten years ago, as the "evidence" the claims were based on was demonstrated to be nothing more than a string of assumptions and coincidences, never backed by any direct proof. 

Just like in the Old West, where mob justice was quick to fall onto the scruffy-looking newcomer as soon as a crime was committed, the innocent hobo spider (Eratigena agrestis) was only guilty of being at the wrong place, at the wrong time. A suspicious appearance, a mistranslated name, and a few unfortunate circumstances were enough to build a nasty but undeserved reputation.

The case of the hobo spider is a textbook example of a "bogeyman" species. It shows how dangerous and misleading it can be to jump to a "spider bite" diagnosis as a go-to explanation for unexplained skin lesions when no spider was seen, and how important it is for members medical and arachnological communities to update themselves and stop perpetuating this misconception.

Who is the hobo spider? 

Our story begins in the early 20th century, probably in the 1930s, somewhere in a port of the Puget Sound, Washington. This is where a spider species from Europe, certainly accidentally transported by ship, was first observed. That species was Eratigena agrestis, the hobo spider. 

Eratigena agrestis, the hobo spider

Eratigena agrestis is a member of the funnel-web spider family Agelenidae (not to be confused with the much more famous and completely unrelated Australian funnel-web spiders, Atracidae).
While not as large as its close relatives, the giant house spiders, the hobo spider is still fairly large, reaching 15 mm in body length and up to 8 cm (about 3 inches) in legspan. Its long, strong legs are straw-coloured to blackish, with spines and long, delicate hairs. The carapace is straw-coloured with darker bands, and the abdomen is grey, with pale chevrons (sometimes quite indistinct). Like most Agelenidae, it has fairly long, pointy spinnerets. The eight eyes are fairly small and all of similar size, in two parallel, slightly procurved (the lateral eyes are further forwards than the median eyes) eye rows. 

In the parts of North America where they occur, giant house spiders (Eratigena atrica and E. duellica, two extremely similar species which can't really be told apart visually), also hitchhikers from Europe, are often confused with the hobo spider. However, they are generally darker, larger on average, and have longer legs. North American grass spiders (genus Agelenopsis), which are part of the same family (Agelenidae) are also frequently mistaken for hobo spiders, but have longer spinnerets, striped legs and different, more contrasted markings.

The hobo spider's legs and carapace are generally quite pale (although the legs can be blackish), and the abdomen is grey with chevron markings (which can be faded)

Giant house spiders (Eratigena group atrica) are usually darker, longer legged and, on average, larger than the hobo spider

In Eurasia, the hobo spider is an extremely widespread species. While it is more common in the southern parts of the continent, it is found all over Europe, and as far East as central Asia. Unlike the giant house spiders, it is not commonly found in close vicinity to humans in its native range; it is generally found under stones and debris in open habitat, such as grassland, fields, road verges, brownfield sites, scrub, or rocky, open woodland.
In North America, on the other hand, it is somewhat more commonly found around man-made constructions, possibly because of lower competition with its larger cousins, the giant house spiders, which are much more limited in range. It is, however, not a true house spider: even though it occasionally wander into houses, it prefers living outside, around buildings, or in unheated storage facilities (barns, warehouses...).

From its landing in the Puget Sound, the hobo spider gradually expanded its range throughout the 20th century, first to the US states bordering Washington and British Columbia in Canada, then continued spreading eastwards. At the end of the century, its range extended from the Pacific coast to the states of Wyoming and Montana, and southwards to the north of Utah. It is now widespread in Utah, as well as in Colorado, and has been found in Nevada and northern California.
In Canada, it occurs in the southern halves of British Columbia and Alberta, as well as in the Canadian part of the Great Lakes region, around Toronto and Ottawa (probably as a result of a different introduction event, or accidental transport from the West).

Because it is frequently found around barns, garages, warehouses and other storage buildings in North America, the hobo spider travels often and easily with humans and goods, hence its common name. 

While Europeans are familiar with Eratigena species, which are a common sight in houses, attics and garages, the hobo spider stood out in its newly conquered range. The giant house spiders of the atrica group also occur in North America, but their distribution is much more limited, and largely restricted to the Pacific northwest (mainly Washington and Oregon and southwestern Canada, with another population in Nova Scotia, and a few scattered records in other places).
Their close cousin, the barn funnel-web spider (Tegenaria domestica) is found all over the continent; it is, however, much smaller and less impressive-looking than the hobo spider. Grass spiders (genus Agelenopsis) are close in size and build similar webs, but do not tend to favour human dwellings as their habitat.
Thus, the hobo spider's conquest of the West did not go unnoticed; many were understandably wary about this impressive-looking newcomer and its large, conspicuous funnel-webs.

Things go south

The wariness towards this species only got worse when some clueless (or ill-intentioned) entomologist coined a new, unfortunate common name for it: "aggressive house spider".
Disowned by the arachnological community, this scary-sounding common name is actually a mistranslation of its scientific name: "agrestis" means "rural", "from the fields", which describes its habitat preferences in Europe.
The fact that it happens to sound like "aggressive" is only a coincidence, and couldn't be further from reality. Just like other members of the genus Eratigena, the hobo spider is not aggressive. It does not bite readily, and will react to disturbance by fleeing or staying still, curled up into a ball.
To get bitten by a hobo spider, one would have to (accidentally or intentionally) press it against their skin; and even like that, it wouldn't systematically try to bite, nor would its fangs necessarily manage to pierce the skin.

The hobo spider does not react to threats by confronting them; it will either flee or curl up into a ball and stay still

Unfortunately, the sinister aura created by this misleading common name undoubtedly played a major part in the mishaps that befell our poor spider in the second half of the 20th century.
However, the hobo spider's misfortune started long before this problematic name was coined, with a series of events which, at first, had nothing to do with it. 

In 1937, while the hobo was making its first steps on the New Continent, Loxosceles laeta, the Chilean recluse, was identified as a causative agent of skin necrosis in Chile. This was the first time a clear, direct correlation was established between a spider's bite and cutaneous ulcers.
20 years later, in 1957, a similar case was observed in the United States, implicating Loxosceles reclusa, the brown recluse. In the following years, North American medical professionals became acquainted with necrotic arachnidism, i.e. spider bites resulting in necrotic ulcers. 

This was the beginning of a widespread and long-standing psychosis involving the brown recluse. The thing with necrotic arachnidism is that it can be extremely difficult to diagnose, as many unrelated conditions and pathogenic agents can cause lesions that look visually indistinguishable from recluse bites, and are very often mistaken for them. As it was, then, a novel and still poorly-known medical entity, "brown recluse bite" quickly became a convenient explanation for any skin lesion of uncertain origin.

Since recluse spiders (genus Loxosceles) were discovered to be able to cause necrotic skin ulcers, they have been widely used as a scapegoat for unexplained skin lesions

In the 1960s, numerous cases of necrotic skin ulcers are diagnosed as recluse bites, regardless of whether or not a spider was actually seen.

Nowadays, even though erroneous identification of lesions as spider bites is still a persisting and important medical problem, a lot of biomedical research in the last 25 years has been focused on improving diagnosis accuracy. Medical practitioners are urged to exercise caution and carefully consider other possibilities when diagnosing potential recluse bites, as summed up by the "Not recluse!" mnemonic device. Actually having seen a spider is still the only criterion that allows positive identification of a lesion as a spider bite (an ELISA test has been developed for detecting recluse venom toxins in tissue, but, after 20 years, it is still not commercially available).
In the absence of this pivotal element of proof, the best one can do is establish a diagnosis where a recluse bite is the most likely explanation left, by ruling out every testable alternative and making sure all of the criteria consistent with a recluse bite are here (starting with the most basic: the suspected bite occurred within Loxosceles' range).

In the years following the discovery of necrotic arachnidism, though, medical professionals were not that careful, and their method was generally quite the opposite of what it is today : their approach was "unless there is a more obvious explanation, it's a spider bite".
What arachnologists and medical professionals know today about recluse bites had yet to be learned from their mistakes.

Because of these hasty diagnoses, an abundance of "recluse bite" cases were described within, but also often outside the range of North American Loxosceles species (which includes the Mississippi basin, where the brown recluse occurs, and the arid southwestern states for the few other species) in the following decades.
In the 1970s and 80s, ulcers were commonly diagnosed as spider bites in places like the Pacific Northwest and Canada, where no species of Loxosceles occurs.

Of course, the scenario of bites from accidentally transported recluse spiders, a shaky (see why in the "bonus" section at the end) and, even today, overused explanation, was commonly invoked to explain these out-of-range "bites".
However, even the wildest conjectures involving this hypothesis couldn't explain the sheer number of cases of skin necrosis in the Pacific northwest.

The real reason was, of course, that these lesions were not spider bites at all. Even inside their range, only a minority of skin necrosis cases are caused by recluse bites. Other pathogens, some of which very common (particularly bacterial infections) are responsible for most of the lesions falsely blamed on spider bites, and, unlike recluse spiders, are not restricted to specific geographical areas. 

Even within places where they are particularly abundant, recluse spiders are actually not a common cause of skin necrosis; other causative agents, such as bacteria and circulatory issues, remain far more prevalent

 

For the 1980s scientists who investigated these cases, though, it had to be a spider, because none of that was known just yet. Therefore, if no recluse occurred in the area, it could only mean that there was another species whose venom caused skin necrosis. 
Enter  Eratigena agrestis: with its large size, recent spread, fearsome appearance, and unfortunate common name, the "aggressive house spider" was the ideal suspect. 

Thus, the hobo spider found itself under investigation, and not a very meticulous one. The method regarded, at the time, as standard for testing a spider species for necrotic venom was to have it experimentally bite lab rabbits, and observe the skin reactions.
Jackpot: in a 1987 study led on 9 rabbits, 5 of them, including the four bitten by male spiders, developed cutaneous necrosis at the bite site.

Add to that the fact that the hobo spider was consistently found in the residence of people with unexplained necrotic lesions, as it was "easily the most common large spider found
around human habitations and [...] likely present on the premises of every residence in
these cities [the cities in question being in Oregon, Washington and Idaho]", and it was enough to establish Eratigena agrestis as the perfect culprit.

From 1987 onwards, a rich record of cases of cutaneous necrosis implicating the hobo spiders accumulated in the literature. Some were extremely severe; a fatal one was even reported.
However, pretty much all of the evidence was purely circumstantial; diagnoses based, at best, on the presence of the spider in the same home as someone with a lesion, or even simply on the fact that the species was known to occur in the area.
Sometimes, evidence as weak as the victim having stepped on a web and seen a spider nearby, without even feeling a bite, was regarded as enough for attributing with certainty a lesion to the hobo spider.

Direct proof, on the other hand, was conspicuously scarce. In fact, between 1987 and 2004, only one confirmed bite (where the spider was felt biting and immediately recovered, kept, and identified as Eratigena agrestis by a relevant authority) was reported, in 1992.
The victim was a 42 years old woman with antecedents of phlebitis, bitten on the ankle, who developed, over the course of several months, a venous ulcer caused by deep vein thrombosis.

As flimsy as it was, this evidence was regarded as largely sufficient. Despite the discrepancies between the clinical pictures of the reported "bites", and the fact that, even back then, it was well-known that skin necrosis could have many other causes besides spider bites, the hobo spider's medical significance was a widely accepted idea.
Thus, in the late 1980s, this species was included by the CDC in the list of medically significant North American spiders, alongside recluses and widows. 

The 1990s were marked by the fear of the hobo spider; in 1994 alone, no less than 66 cases of skin ulcers were blamed on Eratigena agrestis. 
Yet, in spite of the huge numbers of diagnoses, cases where a spider was actually directly observed were conspicuously missing. 

The wind turns

With all these cases piling up, the scarcity of confirmed bites (bites where a spider was directly seen or felt) became increasingly suspicious. Circumstantial evidence is useful for building coherent and convincing scenarios, but if no direct proof ever comes to support it, it develops an increasingly strong smell of red herring. 

Suspicious, as well, the fact that this originally European spider, while blamed for hundreds of skin necrosis cases in North America, was never implicated in a single one in its native range.
The (arguably far-fetched) hypothesis of a genetic mutation within the American subpopulation that would makes its venom more potent was ruled out in 2001, when a comparative study found no significant difference in venom composition between American and European populations of E. agrestis. 
In fact, the similarities were so strong that the study pinpointed the origin of the American population in England (where no hobo spider has ever caused any ulcer), as the venom of the English specimens was exactly identical to the American ones.

The only confirmed bite case recorded did not prove much either, as the lesion which appeared after the bite was identified as a venous ulcer caused by deep vein thrombosis, of which it is a common complication. As the patient had a history of similar circulatory issues, she was at risk of such complications, which were as likely, if not more, to have been triggered by the mechanical injury of the bite, or by secondary infection, than by the venom itself.

Thus, in the early 2000s, some scientists started questioning, and eventually re-investigating the hobo spider's medical significance. 

The reopening of the hobo's file wasn't an isolated event, but was part of a wider movement, which kicked off a major methodological shift in medical arachnology. 

That shift was initiated in Brazil in 1990, when a huge study re-examined bites by Lycosa erythrognatha, until then regarded as a cause of skin necrosis, so much that an antivenom was produced to treat bites by this species. This new research was based on a protocol that was, at the time, completely new: it only included verified bites, i.e. cases where the spider was seen or felt at the moment of the bite, and identified with certainty.
Among the huge sample of 515 patients, not a single one developed the necrosis this spider was previously assumed to cause, leading to its medical significance being re-evaluated. 

This study became foundational by bringing the importance of verified bites under the spotlight, and highlighting how easily misdiagnosed spider bites can be. It prompted new research, with similar methods, on the other species likewise accused of causing necrotic bites.

In 2003, white-tailed spiders from Australia and New Zealand, were similarly exonerated, with a prospective study gathering 130 verified bite cases, none of them necrotic. 

In 2006, it was the genus Cheiracanthium (yellow/long-legged sac spiders) which was in turn cleared, with prospective study and literature review that showed none of the 59 verified bite cases they gathered from patients and the literature from North America, Eurasia and Australia, resulted in any necrotic ulcer.

Accused of necrotic bites in Europe and America based solely on indirect evidence, Cheiracanthium mildei was exonerated in 2005

 

All three genera had been blamed on the same type of evidence, very much like the hobo spider: confronted with patients presenting unexplained ulcers, doctors suspected spiders, based on the fact that one spider genus, Loxosceles, could cause similar lesions. Testing common local spider species to see if they could be the cause, scientists conducted envenomation experiments on rabbits. If the rabbits presented lesions, the spider was considered capable of causing necrosis, and all similar unexplained ulcers observed on people were blamed on them.

What transpired through these new studies was that rabbits are not a reliable animal model for studying the potential effects of spider bites on humans, and that all species incriminated on this type of indirect evidence need to be re-investigated. 

As the hobo spider's reputation was based on the same type of evidence, its medical significance fell under renewed scrutiny in the 2000s.

To test its cytotoxicity  (ability to destroy cell membranes), its venom's hemolytic potential (ability to destroy blood cells) was measured, and found to be low: 0.62% in males and 0.93% in females, while the brown recluse venom, known and proven to be cytotoxic, had a hemolytic potential of 37% at the same dose. Interestingly, the higher potency of the female's venom compared to the male, also observed in a previous study, directly contradicts the experimental results on rabbits, which developed ulcers more frequently from male spiders.
The same study also found out that hobo spiders were not particularly prone to carrying pathogenic bacteria on their fangs either, ruling out the possibility that the ulcers may be caused by bacteria from the spider's fangs.

In 2014, a prospective study on hobo spider bites was attempted, gathering 33 confirmed spider bite cases in Oregon over a 3 year period. This was, however, a partial failure, as out of all the gathered verified bites, only 1 was caused by a hobo spider.  
This was nonetheless a significant result, as none (including the hobo bite) of the recorded bites resulted in skin ulceration, and the fact that only a single hobo spider bite was reported in three years disproves the previously prevalent  assumption that they were a common cause of bites in the Pacific Northwest. 

Some may object that unlike Lampona or Cheiracanthium, the medical significance of Eratigena agrestis was never positively disproven, as there are only two records of verified bites, one with an ulcer of uncertain aetiology, and one without. 

That claim would, however, be dishonest, as it ignores a large body of cumulative proof:
-  Its venom was found to have very low cytotoxicity, making it hard to explain how it would be capable of causing skin necrosis.
- If it did, it would be the only species of its genus, and even of its whole family, to have medically significant venom. While such a case would not, of course, be impossible, it makes it all the less likely.
- While the hobo spider's range covers almost the entire Northern hemisphere and spans across all of Europe and most of Asia, it has only ever been accused of medically significant bites in North America, while not a single case has ever been reported from its immense native range. 
- Finally, and most importantly, there is no valid reason, in the first place, to suspect it could cause necrotic bites. 
The suspicions were built on purely coincidental evidence, i.e. the fact that the spider was found in or around the dwellings of patients with skin ulcers. As the hobo spider was, reportedly "likely present on the premises of every residence in these cities" (the cities in question being larger cities of the Pacific Northwest like Portland), it doesn't prove anything: look long enough in any house in the area, and you'll find one. The experiments conducted on rabbits are no proof either, as the cases of the other falsely accused species have proven they are not a good proxy for estimating the effects on humans. 
On top of that, while hundreds of cases were reported as long as the observation of a spider wasn't required for establishing a diagnosis, not a single one was recorded after it became the norm!

There is, thus, no valid reason to claim or even suspect that Eratigena agrestis may be medically significant anymore. 
In 2015, this absence of proof led the CDC to remove the hobo spider form their list of species of medical concern. Today, necrotic hobo spider bites are regarded as a medical myth, and most sources have updated their information (more or less thoroughly) in accordance to that.
However, there are unfortunately still places, particularly pest control companies (not all of them, of course) who make direct financial gain from keeping people afraid of spiders, who keep passing on outdated and false claims... 

*Directly (and probably not very skilfully) translated from French: "Ce qu'ils veulent, ce n'est pas la vérité, voyez-vous, c'est la cohérence. Une erreur judiciaire est toujours un chef-d’œuvre de cohérence."

References are integrated into the text of the article; the words in blue are clickable and will redirect you to the sources of the information. 

Except when the source is explicitly mentioned, the photos illustrating this article are mine and are not free to use without permission 

All of the contents of this blog are proudly and painstakingly produced without the use of generative AI.