Where did spiders come from?

February 9, 2018 • 9:30 am

The title question is an evolutionary one: what were the ancestors of today’s spiders?

First, some taxonomy. You should know, if you don’t already, that spiders aren’t insects. They’re in the same phylum, Arthropoda, but then belong to the non-insect subphylum Chelicerata, the class Arachnida, and the order Araneae. But their evolutionary origins have been a bit murky, though we know they’re fairly closely related to the order Thelyphonida (also in the Chelicerata and Arachnida), containing whip scorpions—also called vinegaroons. Thelyphonids look sort of spiderlike, but have tails. Here’s a whip scorpion:

Well, the origins of spiders probably started with a creature that looked something like that, according to a new paper by Bo Wang et al. in Nature Ecology & Evolution (reference below, free access [might require UnPaywall], pdf here). The basis of the paper (and an accompanying paper; see below) is the discovery in Myanmar of two pieces of amber dates at least 100 million years old, each containing what appears to be a member of the same species. And that species sheds some light on the origin of spiders.

The authors name that species Chimerarachne yingi. This creature has a tail like a whip scorpion, but also has two distinctive features (“shared derived traits”, also called “synapomorphies”) that characterize spiders. First, sperm-transferring pedipalps, the anterior appendages that put sperm from the male’s gonads (deposited first on a small “web”) into the female’s reproductive opening (both amber specimens are males). Second, spigot-like spinnerets, the nozzles that squirt silk out of the abdomen. Here’s a ventral (bottom) view of one of the specimens, and its interpretation. The length of the black line in the photo is 1 mm (about 1/25th of an inch), so this creature is pretty small.

The pedipalps are labeled “Pd” at the top, and the four spinnerets are labeled “ALS” (anterior lateral spinneret) and “PLS” (posterior lateral spinneret).  You can also clearly see the tail.

The tail of the whip scorpion, and likely of this specimen, serves not as a stinger (as it does in true scorpions), but as a sensory organ. Also, whip scorpions don’t have specialized pedipalps (nor spinnerets); insemination in that group is achieved by males transferring spermatophores (packets of sperm) using their different, pincer-like pedipalps. (Spider pedipalps have all kinds of other features for transferring sperm, including a “palpal bulb” that holds the sperm taken from the male’s abdomen.) Here’s a spider pedipalp versus a whip scorpion pedipalp:

Spider pedipalp (the dark bits are the complex “palpal bulbs”):

Whip scorpion (simple pincers):

The pedipalp of the new specimen C. yingi is complex, like that of a spider. Here’s a photo of part of the pedipalp, labeled “ta” for “tarsus”; the tarsus is modified for sperm transfer and is not a simple pincer. Scale bars in both photos below are 0.2 mm:

Here are the specimen’s four spinnerets:

I won’t belabor you with the morphological details, most of which are for specialists, but the important part is that we have a creature that is part spider and part whip scorpion.  This leaves us with the big question:

Is this a “missing link” between spiders and whip scorpions? No, it isn’t, if by “missing link” you mean some kind of common ancestor.  Nor is it really a “transitional form” between an ancestral creature with a tail and modern spiders, because already at 100 million years ago we see fully modern-ish spiders. This species and its mosaic relatives existed alongside spiders similar to those we see today, but C. yingi and its relatives probably went extinct without leaving descendants.

What it does show is that there were transitional forms between the morphology of the ancestor, which based on phylogeny probably had tails, and that of modern spiders. In other words, this gives us an idea of what the transitional form might have looked like, even though it wasn’t on the evolutionary route to modern spiders. Here’s the phylogeny the authors came up with, showing the placement of C. yingi as a sister group of spiders (Araneae).

Note that this phylogeny is unclear, as different methods give different results. In fact, I just now discovered there’s another paper in the same issue, by D. Huang et al., (reference below) arguing that this specimen belongs in an extinct sister group of spiders, the “Uraraneidae”, as seen in the phylogeny below (Huang et al. put the specimen in that group rather than occupying its own branch).

(from paper): Consensus cladogram (Bayesian majority rule consensus with support values) showing the position of C. yingi within the Pantetrapulmonata branch of the arachnids, which encompasses the spiders and their closest relatives (see Supplementary Fig. 3 for the full annotated tree). Both Attercopus and Permarachne are incomplete and scoring them individually collapses the tree, thus we combined characters from the two genera into one terminal Uraraneida.

So there’s still some confusion about where this specimen resides on the arachnid family tree. But to some extent it doesn’t matter. What this specimen shows is that at one point there was a mosaic species that had the tail of a whip scorpion (in the “pedipalpi” above) and the palps and spinnerets similar to those of a modern spider.

Both papers show that this creature wasn’t an ancestor of modern spiders because it coexisted with modern tailless spiders. What it gives us is an idea of what the transition from tailed arachnid to tailless spider might have looked like, and, importantly also pushes back in time the evolutionary origin of spinnerets and specialized pedipalps, which could have existed in the common ancestor of Uraraneids and spiders (the node labeled “1” in the phylogeny above.

Here’s a figure from the Huang et al. paper (which I’ve only skimmed), that analyzes the same two specimens as Wang et al. to reach a slightly different phylogenetic conclusion. But this figure is very nice, showing the animal with its complex pedipalps (“b”) and spinnerets (“h”):

(from paper): a, General habitus. b, General view of pedipalps. c, Detail of chelicerae (arrow points at cuticular striae). d, Detail of palpal tarsi. e, Detail of leg III (arrow points to metatarsal trichobothrium). f, Detail of the posterior lateral spinneret with the spigots. The red arrows indicate segments. g, Ventral view showing the clear, broad sternum. h, Detail of the spinnerets, pygidium and proximal portion of the flagellum. Scale bars: 1 mm in a; 0.2 mm in b–f; 0.5 mm in g and h.

_________

Wang, B., J. A. Dunlop, P. A. Selden, R. J. Garwood, W. A. Shear, P. Müller, and X. Lei. 2018. Cretaceous arachnid Chimerarachne yingi gen. et sp. nov. illuminates spider origins. Nature Ecology & Evolution, online; doi:10.1038/s41559-017-0449-3

Huang, D., G. Hormiga, C. Cai, Y. Su, Z. Yin, F. Xia, and G. Giribet. 2018. Origin of spiders and their spinning organs illuminated by mid-Cretaceous amber fossils. Nature Ecology & Evolution, online, doi: 10.1038/s41559-018-0475-9

95 thoughts on “Where did spiders come from?”

    1. Heebie-jeebie inducing. I read the whole post (I’m attempting to read one of the Professor CC’s science posts all the way through per week, with full concentration).

      1. Yes, also that! I quite enjoy the science posts, and try to do the same. The novel The Islanders by Christopher Priest features a terrifying arthropod in an alternate timeline that I found equal parts entrancing and horrifying. It was a fun read.

    2. I was hoping that it wasn’t a three-foot-long rattlesnake to provide a sense of scale 🙂

  8. This is really interesting. So, if I understand, the Chimerarachne yingi is what the transition might have looked like. I wonder if the tail (if it did have one) of the ancestor of modern spiders was for stinging or sensing and what the adaptive significance (if that’s the right phrasing) is for losing it.

    1. There’s no indication in either this fossil or whip scorpions that the tail was anything but a sensory organ. If that was the case in the ancestor, which seems likely given the outgroup, then it was sensory here, too. Who knows why it was lost? One hypothesis is that spiders developed sufficient sensory capacities that they no longer needed it, and it disappeared via selection.

        1. Spiders probably used silk for other purposes — eeg cocoons, the sperm-transfer silk thingie mentioned in the text above, and most likely as a drag-line, before any of them wove proper prey-capture webs.

  11. [I don’t want to let a science post go without comment, so please imagine one of my usual puerilisms here.]

    1. I had to look that up: “Childlike behavior by an adult, especially as indicating a mental disorder. quotations”.

  14. Nor is it really a “transitional form” between an ancestral creature with a tail and modern spiders, because already at 100 million years ago we see fully modern-ish spiders.

    Does that rule out it being an ancestral form? Cannot an ancestral species co-exist with species that radiated from it?

    1. Does that rule out it being an ancestral form? Cannot an ancestral species co-exist with species that radiated from it?

      How do you keep the ancestral species from evolving?

      There’s no problem with having organisms that retain ancestral characters or features living side by side with organisms that don’t. The platypus, for instance, retaining ancestral features that marsupials and eutherians don’t, like egg-laying. And a lot has been learned from bird fossils found in China that lived after birds had evolved, yet many of the bird fossils retain ancestral characters that shed light on avian evolution.

      But no one should mistake later forms for being truly transitional species. They’re not. Platypuses don’t appear to have evolved greatly in their morphology (except the ones that evolved into echidnas, which appears likely), yet they certainly have changed, for example, modern adults don’t have teeth (juveniles do, but shed them), while ancient platypus adults did.

      There are no living fossils, really. At least no one knows how there could be.

      Glen Davidson

      1. Yes, you can’t stop evolution in its tracks, but there are species whose essential forms remain little changed over geological time. (If they have no extant close relatives, they are commonly called living fossils.) I see no reason why such a slow-changing species could not coexist with derivative species, in principle. The parthenogenetic marbled crayfish described yesterday co-exists with other crayfish from which surely it derived.

        More than likely, I am talking through my hat because I am not a biologist.

        1. An essential form unchanged over geological time only tells us what it looks like on the outside. Internally, who knows what changes have taken place? New metabolic pathways, for instance. A longer or shorter gut?

          And if the marbled crayfish is a new species, why are there still monkeys?

          IANAB either.

  15. Where did spiders come from?

    Hell. They came from Hell.

    I’ve been thinking of how I would describe this in terms of spider evolution. I’d say that these are not transitional forms but they are descendants of a transitional form between spiders and whip scorpions that didn’t change much.

    I’m a bit confused by this. The trigonotarbids are a sister group to spiders and these animals but trigonos resembled spiders other than having a segmented abdomen and lacking silk. They didn’t have tails either so they must have independently lost their tails from spiders.
    It would be nice to find a cladogram for the whole group.

  16. Not only is this interesting in giving an idea what spiders’ ancestors might have looked like, but these fossils are stunningly beautifully preserved, about a 100 M years old! That is 35 million years before the demise of the dinosaurs.
    As stunning as that enantiornithean (‘opposite bird’) hatchling of a few days ago!

    1. Might, btw, the evolution and radiation of ‘real’ spiders not be linked to the radiation of insects after the evolution of insect pollinated flowers? The timing appears roughly correct (if I’m not mistaken), and insects are – again roughly- the mainstay of spiders’ diet…

    1. Yes, I think arachnids, terrifying as they are to me, represent a very good body type. Those whip scorpions with the hands are going to develop tool use and terrorize us, I just know it.

  19. This is great to see. One of the coauthors on the paper, W.A. Shear, is William Shear, who is now an emeritus Professor of Biology at my alma mater, Hampden-Sydney College (Virginia). I took several of his classes while I was at Hampden-Sydney, including a great evolutionary theory class. Hampden-Sydney is a tiny liberal arts college (~950 students) and Dr. Shear was an excellent teacher. He kept a container of sand on his desk with antlions in it. So cool to see that he continues his excellent work.

  21. Wow! Lots of arthropody pictures today. Very interesting stuff.
    Since it is mosaic for both spiders and whip scorpions, I am not sure why it is placed closer to the spiders rather than being closer to the whip scorpions.

    1. It’s not really mosaic, as I would understand the term: There is a long list of features in this fossil that ONLY spiders have, that we are sure were lacking in the common ancestor of spiders and whip scorpions. Notably the segmented spinnerets and the sperm-transfer mechanism in the pedipalp, but also the form of the chelicerae.

      It just happens that the flagellum [“whip”] was present in the common ancestor, but lost in the common ancestor of modern spiders. We already had evidence of that from previously known compression fossils. But the beauty of these amber fossils is that all the relevant structures are preserved.

  23. Those amber-preserved specimens are amazing! I hope some lab that works with ancient DNA is successful at sequencing some of C. yingi’s genome. That would be really cool because we could see its placement in a molecular phylogeny as well as in this morphological one.

    1. I was wondering that too. If DNA is readable, that would help locate the animal more precisely – as long as there is enough data on nearby relatives.

          1. (somehow hit post before comment was completed; if you happen to see these, Jerry, please delete this comment and the one directly above…)

  26. I’m curious why scientists are comfortable thinking in terms of transitional forms. It seems to me like saying a given link in a chain is transitional to the other links.

  28. I was expecting the answer, “from the bowels of Hell,” but the actual article was quite fascinating. Thanks!

  29. I believe that while no living examples of this beastie is known, there will be or are already ongoing searches for it based on the chance that it does still exist in the area where the fossil was found, as they are small and the region is poorly studied.

  30. What can an ex-chemist do but gulp in admiration? Thanks so much for the science. Please keep it coming!

  31. The Unpaywall extension couldn’t find any legal open-access version of this article.

    So I had to go to Sci-Hub. NB : UK DNS providers are being forced to delist Sci-Hub, … and they’ve added another level of blocking since I tried last.
    So, no paper for me.
    I recall some years back that people were doing cladistics on the arrangement of mouth parts in the various Arthropod sub-phyla, as an indicator of their original branching from a putative ur-Arthropod. They postulated that a number of (pre-oral?) segments of the head region had fused together to form the cephalothorax body section, with the various mouth parts being an indicator of how many fused segments there were, and the proposal that this was a very deep character. I can’t remember the outcome – but the approach made sense. It’s probably been superseded by genetics data these days, but I should be able to dig the paper out from the piling system, if someone is interested.
    Oh, there’s some murkiness :

    Publisher’s note: Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.

    Sounds like the origin region of the amber is … disputed.

  34. I’m always very interested in new results about spider evolution (I love spiders). The following sentence was particularly interesting: “Both papers show that this creature wasn’t an ancestor of modern spiders because it coexisted with modern tailless spiders”. How watertight is that argument? It strikes me that it should be possible (at least in theory) for the evolutionary tree to recombine, at least at the level of families or maybe even genera. That is to say, it should be possible for one genus to branch into two separate genera, which then give rise to different species in the same subsequent genus. Is there a good theoretical reason why this type of “convergent evolution” cannot occur? If not, is there any evidence that it may have occurred?

  36. This was a mighty fine read.

    Is there any substance other than amber that has the ability to preserve insects like this? I know tar and extreme cold can preserve many animals, but insects? We’re lucky for the trees for their amber time-capsules.

  37. The whip scorpion sprays acid through the flagellum?

    Is the scorpion stinger directly related to the flagellum?

    Do scorpions have structures related to the spinnerettes?

    Trivia : in The Hobbit, Bilbo insults the spiders with the name “attercop”.

    1. 1) Not “through”, but the spigot is at the base of the flagellum.
      2) Maybe, maybe not. The scorpion sting is a modified telson [in arthropods, this is the segment-like structure posterior to the anus [the latter borne on the last true segment. But the sting-bearing “tail” of the scorpion includes 4-5 segments anterior to the telson. The flagellum in whip-scorpions and various other arachnids — now including the ur-spider — just might be a modified telson.
      3) The spinnerets are modified segmental appendages, so serially homologous to the legs. They are present in all spiders, and only spiders. Scorpions have so-called comb-organs [pectines] on the under-surface of the second segment of their mesosoma [“abdomen”]. These sensory structures are also modified limbs, but are not homologous to the spinnerets.

      I have no comment on Tolkien, other than the general common-sense rule that one should not insult something bigger and nastier than yourself..

      1. I see

        Serially homologous to legs – is there a distinct number of spinnerettes? The way there’s always, we know so well, eight legs?

        Follow up about the silk, the acid, the charybdotoxin – these presumably originate from separate organs – do whip scorpions excrete charybdotoxin?

        … and then are there vestigial structures in spiders related to the acid/charybdotoxin?

  38. While this post is quite interesting, I must admit, I hate spiders. Little buggers annoy me no end. So much so, I murder them whenever they bug me.

    I live in a motor home and tow a small car behind when I travel. At one RV park I used to stay at, I parked the car about three feet from a hedge. Whenever I went to the car in the morning, I would find a spider had built a web from the hedge to my car. Sometimes I walked right into the web before spotting it. Uhg. One morning I spotted the web, grabbed a broom from the RV and ran to the car. A spider was sitting in the web. I knocked him to the ground with the broom and stomped on him. Next morning, the web was back. Another spider was sitting in it. As I grabbed the broom, it spotted me and scurried to the safety of the hedgerow. I couldn’t find the spider to kill him so I declared war. Bought a can of spider spray, sprayed the entire hedge row, keeping that up until no more webs found their way to my car. I think I demonstrated that spiders can learn rapidly. That little beast must have seen me kill his buddy the day before and decided to abandon web before the nasty giant with the broom could do the same to him.

    I still vigorously go after spider webs whenever I see them on my RV but occasionally one will find its way inside where I have to deal with it there. Me and spiders don’t get along.

      1. None, that I know of. Spiders returned from time to time and I kept spraying. Of course, everything is gone now. The owner sold the park which was turned into a housing development. But my spraying had nothing to do with that.

    2. Well, occasionally after waking up there’s a big red bump on me. I figure a spider bit me during the night. They get into places. What can you do.

        1. Ants then?

          Ladybugs?

          Stink bugs?

          This problem hasn’t gone away and I haven’t seen any other cutesy-crawlies in my environs…

          1. I think waking up with an apparent bite now and then is not uncommon (it happens to me now and then). The articles that suggest they’re not from spiders offer only such suggestions as fleas & bedbugs, which have definitely never been the case for me! (Well, except for that one time on an OTS course in Costa Rica where we slept overnight on the floor of some dicey inn and all awoke with fleas…)

            Nonetheless, it’s highly unlikely that a spider would get into bedding and bite a person. We are not their prey. Perhaps we’ve received a mosquito/biting midge/biting gnat/sand fly, etc., bite, and our histimine level didn’t reach noticeable strength till we were asleep.

          2. Animals can bite for reasons besides predator-prey relationships.

            Nonetheless, I see your point – perhaps it could be a lone flea, a lone … louse?…. but I thought those come in large numbers or not at all…

            … and I have noticed, this is a Science Post(TM), yet, I diverted it into a Health Post(_?_)…

            is it too early to propose a new law?

            Any science post will, over time, lead to discussions about health?

  42. Though you do occasionally (and rightly) worry over the comparatively low readership of your science commentaries (as opposed to stuff on creationists or politics), make no mistake that your pieces on cutting edge science work, providing context and potential inferences, are most welcome. I have drawn upon them very often in my own TIP work trying to keep abreast of the ongoing science, you’re one of those whose fingers are on the pulse of that work, and so keep up that work.

Comments are closed.