Friday, June 18, 2010

Patio Spider Part 2: Pincer, Fork & Key

So how does a Sowbug Killer get past a Woodlouse’s armor? After the shameless cliffhanger I left you with Wednesday, it had better be a good story. And you know what? It is.

Tangent: I have to break these TdF-type posts up oftentimes because I just can’t steal enough time away in a day to bang the whole thing out. I wish that weren’t the case, because I usually get the idea for a post pretty quickly after I learn about something. But I can’t magically just “spray” the idea onto the screen; I have to piece it together, write text to support it, then build my Expand-O-Graphics. Time, time, time. When you get down to it, the Secret of Life is this: the only thing you are given in life is time. The only thing you have to figure out in life is what to do with the time you have.

The short answer is that there are multiple ways to kill a Woodlouse, and the method employed depends on the particular species of Dysdera spider. Which brings us back to species.

The Amazing Story of Dysderid Evolution

In Part 1 I mentioned in passing that D. crocata is 1 of more than 200 species of Dysdera, and that it’s the only one with a worldwide distribution. All the rest are limited to a “Greater Mediterranean” region, which includes several island chains off the West coast of North Africa. That in itself is interesting enough- 200+ species, only 1 outside of the home region- and that one is all over the world. Far out- eh?

But it gets even weirder when you look at their distribution across those Atlantic island chains.

NA island groups map Of those 200+ species, 5 are endemic- that is native to no place else in the world- to the Madeira Islands. 1 species is endemic to the Azores, another 1 endemic to the Cape Verde Islands, and 1 more to Selvagens Islands. Guess how many species are endemic to the Canary Islands? 43. that’s right- out of 200+ Dysdera species in the world, 43 of them are endemic to the Canaries, an archipelago of 10 islands in the Atlantic Ocean about 60 miles off the coast of Morocco.

Side Note: The Canaries are interesting for lots of other reasons. Canary Island flora has a few other head-scratchers like the Dysdera thing, and the archipelago is loaded with dozens of species of endemic snails, beetles and millipedes. map_of_canary-islands And interestingly, for a group of islands that’s been known and studied for centuries, geologists still can’t quite agree on the exact processes and forces behind its origins. Most major islands have a pretty well-agreed upon modern origins story: the Hawaiian Islands are the result of “hot spot” volcanism. New Caledonia is a fragment of an ancient supercontinent. But the origin of the Canaries, while clearly the result of volcanic activity, is still controversial. Like Hawaii, there’s a general age-directional trend: the Eastern islands are the oldest, dating back as long as 20 million years ago, while more Westerly Tenerife is probably only around 2 million years old, But then confusingly it appears that even more-Westerly La Gomera is older than Tenerife, breaking the age-directional trend. And even more confusingly, the eruptions that created Tenerife joined 3 pre-existing islands which date back 6 – 16 million years. Westernmost El Hierro is less than a million years old. It’s suspected that the archipelago’s unique origins may be due in part to its being so close to a passive continental plate (Africa). In any case, it’s a complicated place.

It appears that the Canaries have been a secondary center of Dysdera of evolution in the same way that Mexico may have been a secondary center of Pinus evolution- some pioneers got there, then speciated and radiated. The endemic Dysderids of the Selvagens, the Cape Verdes and (possibly) the Madeiras all appear to be descended from Canary-based ancestors. (The Azores species looks likelier to be the result of a separate colonization from the mainland.)

NA island groups colonization map But the Canary-Dysdera story is maybe even more interesting than the Mexico-Pinus story in that it is almost certainly the result of multiple colonizations. 36 of the 43 endemics look to be descended from a common colonizing ancestor, but the remaining 7 species are trickier to figure out, and appear to be the result of between 1 and 3 separate colonizations.

Extra Detail: An interesting question is how these original colonists got to the Canaries, as they arrived long before people were sailing around in boats. Many juvenile and/or smaller species of spiders travel long distances via “ballooning”, where they trail strands of silk that are caught by wind currents, lifting and carrying them through the air. But no Dysderid has ever been known to balloon. A more plausible dispersal mechanism would be rafting via seaborne mats/clumps of vegetation/debris/soil washed out to sea. Such mats/”floating islands” often emerge from the mouths of rivers in rainy season.

But the rivers that flow out to Morocco’s Atlantic coast, which pass through arid, relatively low-vegetation areas, don’t produce such “floating islands” of vegetation/debris/soil. But in the not-too-distant past, Morocco had a much different climate- cooler, wetter and more forested, and some of the river valleys have topologies which suggest much greater, more powerful flood flows in times past.

Wow. What a totally cool story. But what already does any of this have to do with eating woodlice??

How To Kill A Woodlouse

I’m getting there. Now it turns out that these 200+ Dysderids, though overwhelmingly running hunters, have adapted to various hunting techniques, environments and lifestyles*, including both choice of prey and method of attack, and some of those adaptations can be seen in their chelicerae (fangs)**.

*A few Dysdera species are partially or completely troglodytic. D. unguimmanis, which is native to Tenerife, is unique in the genus in that it has no eyes. It spends its days deep inside dark ocean caves.

**Spider cheliceraea are thought to also vary in some species as a result of sexual selection. But in all known Dysdera species, chelicerae morphology is the same across age and gender, so it’s thought not to be the case with these guys.

To review, chelicerae- which we looked at when discussing Tarantulas in the Oquirrh Mountains- are mouthpart/appendages used by arachnids to grasp and handle food; in spiders (order Araneae) they’ve evolved into hollow venom-injecting fangs. Spider chelicerae have 2 segments- the base, and the fang. Across Dysdera species the form and length of both segments vary considerably and seem directly related to both choice of prey and method of attack.

ExpandO Fang Bogus Wow, what a great Expand-O-Graphic! I’ve really taken things up to a new level, haven’t I? Haha! Just kidding! The inset is a photo I copied from here on No, I didn’t have a fang photo anywhere near that good, in part because I am a lame photographer, but more to the point (and in my defense) because when I photographed the spider I had no idea what it was, nor that its fang morphology would be so important to the story. In any case, here’s a (lame and blurry) non-cheater graphic.

Fang Pointer Specifically, there are 3 primary attack modes used by Dysderids against Woodlice, along with a number of species-specific variations.


The first method is the Pincer tactic, which is used by those species with the longest fangs, and is the method used by our own local SKiller, D. crocata. These guys run quickly up to a woodlouse, turn their “heads*” sideways and pince the woodlouse with their fangs. The upper fang doesn’t pierce the prey; it just provides leverage against the woodlouse’s armored back. The lower Spider Partsfang pierces the soft underside, enabling venom injection.

*Spiders don’t really have “heads” of course. What they pivot is their cephalothorax, one of the 2 body segments of a spider, the other being the abdomen.

Now this doesn’t always work right away; sometimes the woodlouse succeeds in rolling up in a ball first. When this happens different species try different tricks. Some will pick up the “ball” with their forelegs and turn it over and around repeatedly, searching for a gap between the plates into which they can work a fang. But others will simply sit and wait. After a short while- usually less than a minute- an undisturbed woodlouse will typically unroll, enabling another attack.

Pincer tactic view Extra Detail: Say what? How dumb is a woodlouse? The spider is sitting right there! Why does it unroll? The key is most likely lack of motion. Remember last Fall when we looked at compound eyes, I mentioned that while compound eyes are poorer than camera-style eyes for image-resolution, they’re great at detecting motion. A Woodlouse has very simple compound eyes, with only 15-50 ommatidia each, meaning that its image-resolution must totally suck. If a spider doesn’t move, a woodlouse can’t see it.

It gets even better- the species D. abdominalis gently taps a woodlouse with its front legs before an attack, which seems to somehow calm the creature and dissuade it from rolling up. It’s like a hypnotic vampire attack from a horror movie!

Pincer tactic view VENTRAL Extra Detail/Tangent: Speaking of horror movies, one of the rather yucky things you may already know about spiders in general is that they only eat liquid food, and so digest their food prior to consuming it, a task which accomplished by injecting their prey with digestive juices. For a while this bothered me, but upon further reflection I’ve come to admire it, in the same way I admire lightweight, low-impact, sleep-under-the-stars campers. Hear me out.

When you read various sources about spiders, you’ll often read something to the effect of, “Spiders can’t eat solid food…” making their habit of injecting digestive juices into their prey to liquefy it sound like a way of compensating for a handicap- the lack of an extensive internal digestive system. But when you think about it, it’s not a handicap, but an advantage.

Large carnivores like us and mountain lions* and wolves consume mainly vertebrate prey which we must eat from the outside-in, because the flesh is on the outside, and the bones on the inside. But spiders eat invertebrate prey which must be eaten from the inside-out, because the bones are on the outside. Now if your food is nicely contained in a hard, (more or less) watertight exoskeletal container, why not digest it right there, instead of growing, maintaining and lugging around a large, complicated, metabolic-intensive internal digestive system which is prone to various distress, discomfort, blockages, cancers, etc.? We unfortunately don’t have that luxury, because our meat isn’t nicely packaged in convenient containers (and also because we’re not strict carnivores) and so we lug around all this extra, complicated, problem-prone equipment. We’re like campers who drag those big-ass RV/motorhomes all over the place, while spiders are like lightweight, lie-down-on-the-ground-not-a-care-in-the-world-Excellent-Campers. Spiders are way cool.

*I finally stopped saying “cougar”. Just got tired of all the jokes. I thought the slang was dumb and kept hoping it would pass by, like ”Don’t go there…”, or “You go girl!…” but unfortunately it seems to be sticking around.


The second method is the Fork tactic, which is employed by Dysderids with basal chelicerae segments that are concave on the upward-facing/top side.

Fork tactic view These spiders run up to the woodlouse quickly, grasp it up top with their forelegs, and stab upward from below with their up-turned fangs, piercing the soft underside with not one but both fangs. The concave upper basal surface helps to position the edge of the woodlouse’s dorsal armor on top of them.

Fork tactic view VENTRAL This tactic only works if the spider attacks before the woodlouse rolls up.


The third method is the so-called Key tactic, used by Dysderids with fangs that are thin, flat, and fairly elastic. In this method the spider attacks not the soft underside, but the heavily armored back of the rolled-up woodlouse, probing with a single fang until it is able to work it into a tiny gap between two of the armored plates, like a key into a lock.

Key tactic view All of these tactics correspond to the morphology of the mouthparts of the species in question, which brings us to prey preference, as not all Dysderid species eat woodlice all the time, or, in some cases, at all. The species can be broadly lumped into 3 categories. First is those species with un-modified chelicerae- not flat, not long, just “regular”. Many of these species ignore woodlice altogether (though they effectively hunt other invertebrates). At the other end of the spectrum are obligatory, or near-obligatory woodlice specialists, who eat woodlice and practically nothing else, ignoring other prey when offered in experiments.

And then in between these 2 extremes are species that effectively hunt woodlice, but happily hunt other prey as well. Dcrocata These “Flexibles” occupy a middle ecological ground, and at the top of the list of the Flexibles is, that’s right, SKiller/D. crocata. SKillers in laboratory experiments, while effective woodlouse hunters, did not show any preference for them over other prey, happily attacking anything small and slow enough for them to catch. Biological flexibility, or “ecological plasticity”, is a common characteristic of many of the world’s most successful colonizers, from pigeons and starlings to crabgrass and dandelions to paper wasps and rats*, and is likely one of the reasons why D. crocata, alone of the Dysderids, has conquered the rest of the world.

*I so need to do a post on rats.

Pretty cool story for a spider on a patio.

Note about sources: Special thanks to my friend and fellow nature-blogger KB for research assistance. Info about distribution of Dysdera species and the phylogeny and natural history of Dysderids in the Canary Islands came from this paper. Info on the woodlouse-hunting tactics and Dysderid chelicerae morphology came from this paper. Info on the eyes of woodlice came from Lander University’s Invertebrate Anatomy Online site. D. crocata fang-zoom photo inset came from


KanyonKris said...

I'm glad you posted the final today. It would have been torture to wait over the weekend. Don't worry about splitting up posts, it worked fine and that would have been a looong single post.

Woodlouse will consistently unroll in my hand after a while, so it's easy to believe would do unroll with a still spider nearby.

Lots of cool details in this post. So many amazing stories all around us.

ElGuapo said...

I'm surprised these spiders aren't more common where you are. I see them all the time here in Stansbury Park. Seems like every time you pick up a rock or move a log they're there. Also spotted several in our unfinished basement.

Oddly enough these rather ugly red spiders never seem frightening to me like other spiders. I thought of them as slow-moving, but maybe I've just never seen one in action. I assume they're not dangerous to humans?

And speaking of which, do sowbugs cause humans any harm? I mean they're everywhere but is there any reason to even try to control them? I now think of them as little lobsters instead of bugs. Luckily I'm not into eating lobster.

Anonymous said...

Hey Watcher,
I was taking pictures of a tree, like a week ago. you mentioned botany.

If you see any interesting or mysterious plants, or other organisms on your travels, let me know...
-Search for "Greenspire Tree"

Watcher said...

El Guapo- No, they’re harmless. The bite hurts, but it doesn’t do any real damage. Maybe not scary because not hairy? Woodlice: totally harmless- don’t spread disease, don’t eat our food, or wood. I actually love lobster, and try not to think of them as big bugs… BTW, here’s my theory on lobsters: if they could squeal like a cat or a dog, no one would ever manage to get them into a pot.

Gardengeek- Hey thanks for stopping by and thanks for the pointer. Was great running into you last week and I’ll be spending some time over on the site!

Lucy said...

Those red-orange spiders totally gross me out. Other spiders - no problem. I think it is because their abdomens look like big fat wood tics. I was hoping they were a nuisance to humans, so I wouldn't feel bad if I killed one on instant reaction to their hideousness, but now I have to be careful and nice. I hate pill bugs too, so in that respect I should make friends with the ugly orange SKillers and let them do their thing. At least now with knowledge of their great biology story thanks to this blog, I can appreciate their evolutionary prowess while being grossed out.