Remarkable Things

Yesterday was indeed a day for remarkable things. It started when I walked into the bathroom and found a winged male embiopteran sitting on the wall above the toilet cistern. Not the usual place where one would expect to find an embiopteran. It continued when I was informed of the publication of not one but two papers of note on harvestmen.


Lateral view of the male of Neopantopsalis thaumatopoios with most of the legs removed (because there's a limit to how much I'm willing to draw). Even for harvestmen, species of Neopantopsalis have stupidly long appendages. This figure was used in Taylor & Hunt (2009). The scale bar equals a millimetre.

The first is of note on a more personal level - my paper on the new genus Neopantopsalis has come out in Zootaxa (Taylor & Hunt, 2009). This is the first of the three major papers that will be coming out of my PhD - the other two (which I'm still in the process of writing) will cover the genera Megalopsalis and Spinicrus, respectively, as well as the phylogeny of the family Monoscutidae as a whole (or, at least, as much of it as I can reliably make out - long-legged harvestmen aren't exactly brimming over with phylogenetically useful characters). But now that the Neopantopsalis paper has completed the review process and made it into print, I can safely say that I'm not entirely happy with it. I began working on it when I found that Glenn Hunt, the last major worker on Australian harvestmen, had designated a specimen in the Queensland Museum as the type of a new species that he had unfortunately never published before he passed away (in recognition of this, I included Glenn as a second author on my manuscript). This species, it turned out, was one of a well-defined group of species found in Queensland and northern New South Wales - the group now labelled by the name Neopantopsalis. Unfortunately, distinguishing individual species within Neopantopsalis threatened to become an overwhelming task. Individual variation in almost every character was the norm rather than the exception, and in more than one case I was left at a loss to decipher whether I was dealing with a species complex or a complex species. Eventually realising that I could end up spending my entire doctorate working on this one genus (which would not be ideal), I was forced to cut my losses, pick out some well-defined exemplars that could stand in for the overall diversity, and put an appropriate manuscript together as best I could so I could move on to the next topic. So if anyone with a penchant for arachnid systematics with connections to the Queensland region is ever looking for something to do, there's a very widespread genus there still begging to be given the attention it really requires.


The fossil remains of Mesobunus martensi, the better-preserved of the two recent finds. Figure from Huang et al. (in press).

The other paper I learnt of yesterday has perhaps got a broader appeal - the first Jurassic harvestman fossils (Huang, Selden & Dunlop, in press, 2009). Harvestmen are purely terrestrial, not particularly vagile and mostly very delicately built. As a result, their fossil record can only be described as pitiful. There's a couple of fossils from the Devonian, a small collection from the Carboniferous, a couple from the Cretaceous and a small smattering from the Cenozoic (mostly amber) (Dunlop, 2007). However, the few fossils that we do have are quite remarkable in light of how incredibly unremarkable they are. Most fossil harvestmen are almost indistinguishable from taxa living today. Even the very oldest known harvestman, Eophalangium sheari from the Rhynie Chert, would probably fail to raise a single eyebrow if reanimated and released into the modern environent. The origin of harvestmen could not have been all that long before the time of Rhynie Chert, because it wasn't that long before then that there wasn't even a terrestrial environment for there to be harvestmen in. Harvestmen, it seems, are the ultimate retroactive conservatives - if it was good enough 400 million years ago, it's good enough today.

The two new Jurassic fossils, coming from Daohuguo in China, do not buck this trend in the least. Indeed, so similar to modern long-legged harvestmen are they that Huang et al. even assign the better-preserved of the two, Mesobunus martensi, to the modern family Sclerosomatidae (this is the family that includes the genera Gagrella and Leiobunum). They do this on the basis of "the extremely elongate legs, a single tarsal apotele [claw], a pediform [leg-like] pedipalp, and, particularly, the fusion of the first five opisthosomal tergites into a single dorsal plate". They also note the presence of what may (or may not) be pseudoarticulations in femora of the fourth pair of legs, which (if present) would not only place Mesobunus in the Sclerosomatidae, but also within the subfamily Gagrellinae within the Sclerosomatidae. The long legs, simple claw and leg-like pedipalps are plesiomorphies for the harvestman superfamily Phalangioidea (and quite possibly for a larger subgroup of harvestmen), so are not really significant. The dorsal scute and possible pseudoarticulations are more interesting - but, unfortunately, not conclusive. A similar dorsal scute is found in other harvestmen in other suborders (such as the genus Ischyropsalis), and also (though less sclerotised) in some members of the eupnoan family Monoscutidae (notably the genus *ahem*, which however doesn't have very long legs*). Femoral pseudoarticulations are also not unique to Sclerosomatidae - for instance, they have recently been recorded in two species of Monoscutidae by - oh, will you look at that - Taylor & Hunt (2009). So while it is true that Sclerosomatidae is the only living family that shows the exact combination of characters seen in Mesobunus, none of the characters individually is unique. Or to turn that around, while I'm not entirely convinced that Mesobunus is a sclerosomatid, I can't exactly show that it's not a sclerosomatid either. Unfortunately, a rock-solid identification with Sclerosomatidae would probably require examination of features such as the genitalia or spiracles - both highly unlikely to be visible in a fossil.

*I say "ahem" because it's in press as we speak.


Reconstructions of Mesobunus from Huang et al. (in press).

If we accept for the present that Mesobunus is a sclerosomatid, that has some interesting implications for harvestman biogeography. The Phalangioidea can be roughly divided into two morphological groups, a mostly Northern Hemisphere group containing the families Phalangiidae and Sclerosomatidae, and a Southern Hemisphere group containing the families Neopilionidae and Monoscutidae*. The Phalangiidae + Sclerosomatidae group is well supported by a few good characters (most notably the structure of the spiracle) and is more than likely a good clade. The characters uniting the Southern Hemisphere families, on the other hand, are probably plesiomorphies, so this group is quite possibly paraphyletic with regard to the Northern clade (though exact relationships are currently unknown). Similar patterns, with Northern Hemisphere taxa nested among Southern Hemisphere taxa, have been observed in many groups of organisms, and it has often been suggested to indicate a Gondwanan ancestry for those groups. Birds, butterflies... there was a period when it seemed almost everything came from Gondwana. Usually, such Gondwanan ancestries were suggested to be related to the mass extinction at the end of the Cretaceous - with the Northern Hemisphere more heavily affected by the end-Cretaceous meteor impact than the South (which is entirely plausible if the Chicxulub crater in Mexico is the site of the impact), Southern Hemisphere taxa were able to radiate at the beginning of the Cenozoic and repopulate the devastated Northern Hemisphere.

*Phalangiidae extend into the Southern Hemisphere in Africa, and Sclerosomatidae in South America, but in both cases it seems likely that these are more recent invasions from Northern Hemisphere ancestors.

The problem with a Gondwanan origin for the Phalangioidea, however, is that it implies a rather recent derivation for the Northern Hemisphere clade, within the last hundred million years or so, which seems a little out of kilter with the sedate rate of harvestman evolution suggested by the fossil record (most Eocene amber fossils [including Phalangioidea], for instance, can be assigned not only to modern families but even to modern genera). On the other hand, if sclerosomatids were present in the Middle Jurassic of China as suggested by Mesobunus, then that indicates that the modern phalangioid families had diverged before Gondwana had even properly divided from the rest of Pangaea, and some other explanation is required for modern phalangioid distribution.

REFERENCES

Dunlop, J. A. 2007. Paleontology. In Harvestmen: The Biology of Opiliones (R. Pinto-da-Rocha, G. Machado & G. Giribet, eds) pp. 247-265. Harvard University Press: Cambridge (Massachusetts).

Huang, D., P. A. Selden & J. A. Dunlop (in press, 2009). Harvestmen (Arachnida: Opiliones) from the Middle Jurassic of China. Naturwissenschaften.

Taylor, C. K., & G. S. Hunt. 2009. New genus of Megalopsalidinae (Arachnida: Opiliones: Monoscutidae) from north-eastern Australia. Zootaxa 2130: 41-59.