Osgood, Fuertes, and mice that swim and mice that wade

The manuscript on Crato turtles has been completed and submitted, the Galve vertebrates manuscript has been completed (to all intent and purposes), and the review of IUP’s The Carnivorous Dinosaurs has been completed (again, to all intent and purposes). So, finally, I’m hard at work on the phd, and once more deeply immersed in the world of basal tyrannosauroids. But having spent the better part of the day coding characters I think I will allow myself some blog time. If you think I’m going to post about Guanlong, Dilong, or even Eotyrannus, well: you don’t know me very well do you?

If you like the idea of being steeped in the lore of natural history research, then the literature on African amphibious murids provides rich pickings. Take the discovery of the obscure Ethiopian mouse Nilopegamys plumbeus, collected in 1927 by a field assistant of Wilfred H. Osgood on a tributary of the Blue Nile in north-eastern Ethiopia. Osgood (1875-1947) gained his reputation as an ornithologist and specialized during the 1890s in oology, but in 1897 he joined the then US Bureau of Economic Ornithology and Mammalogy (later to become the US Biological Survey) and embarked on significant collecting trips to California and Alaska. He later founded the Cooper Ornithological Club, and in 1909 joined the Field Museum of Natural History in Chicago. It was while based there that he made his best-known contributions: those published during the 1920s and 30s on the mammals of Africa (particularly Ethiopia) and South America (particularly Chile), and in particular on the rodents. In 1927, Osgood took part in an expedition jointly funded by the Field Museum and the Chicago Daily News.

In the field with Osgood was Louis Agassiz Fuertes (1874-1927), one of the most talented and revered of late 19th/early 20th century natural history artists (though he wasn’t just an artist, as he also lectured). Predominantly interested in birds, Fuertes - like Osgood - had explored Alaska in the late 1890s but later traveled widely across the Americas and Africa. He illustrated countless books, magazines and museum murals. And, yes, he was named after the Harvard professor and naturalist Louis Agassiz.

During that 1927 field trip, it was Fuertes’ job to draw the specimens obtained by Osgood’s party. Presented with the new rodent later named Nilopegamys, the sketch Fuertes produced is the only illustration that depicts a fresh specimen. It was also one of his last illustrations because, on returning home to the USA in August of that year, he was killed when a train struck his car at Potter’s Crossing, Unadilla (New York). The illustrations in the car at the time – which included the Nilopegamys sketch – were thrown from the vehicle during the collision.

Osgood (1928) described Nilopegamys as an entirely new sort of murid for Africa: as an amphibious swimmer most like the South American fish-eating rats Ichthyomys. But Nilopegamys was clearly not as specialized for amphibious life as are the ichthyomyines, and furthermore Osgood’s description was brief and without thorough comparisons to some other tropical African murids. Consequently it was suggested during the 1960s that Nilopegamys wasn’t a distinct taxon, and that it should be sunk into synonymy with Colomys goslingi, the Velvet rat. A long-limbed murid with an impressive array of whiskers, Colomys is amphibious and hunts for arthropods, worms and molluscs along stream and swamp edges. The consensus opinion became that Osgood and Fuertes had been incorrect about the validity of Nilopegamys, and, in time, it disappeared from the textbooks.

But it turns out that this decision was rash. Redescribing Nilopegamys in 1995, Julian Kerbis Peterhans and Bruce Patterson showed that Nilopegamys was clearly morphologically distinct from Colomys, and certainly worthy of generic recognition. While both genera are similar in their velvety fur and sharp demarcation between dark upperside and white underside, they differ in that Nilopegamys is larger, with broader feet that possess hairy margins, and with proportionally smaller ear pinnae. The two also differ in the arrangement of pads on their feet, in the number of roots their molars have, in the sizes of their foramen magnum, and in other details (Kerbis Peterhans & Patterson 1995). The features that distinguish Nilopegamys from Colomys suggest that it is more specialized for aquatic life than Colomys is. In essence, it seems to be evolving toward an ichthyomyine-like condition, and it certainly possesses several of the characters that Voss (1988) listed as being correlated with amphibious habits in murids (dense and soft fur, enlarged hind feet, enlarged braincase, reduced visual and olfactory senses etc.).

Nilopegamys and Colomys are both different from another tropical African form, Malacomys: the long-footed rats, big-eared swamp rats or long-eared marsh rats. Poorly known, but apparently ranging throughout most of central Africa, Malacomys looks like a mouse on stilts, at least when it’s not crouching. Because these amphibious mice differ in so many of their anatomical details (and share few detailed, uniquely derived characters), Kerbis Peterhans & Patterson (1995) suggested that the similarities apparent between them are due to convergence. It would be nice, however, to test this by plugging them into a phylogeny. However, people are only just starting to work on parsimony-based murid phylogenies, and I’m not aware of any that incorporate Nilopegamys or Colomys. Malacomys at least appears to represent a distinct lineage within the so-called core murine clade, not too distant phylogenetically from Mus and Apodemus (Steppan et al. 2005), but there are suggestions that it is not monophyletic and that two distinct lineages may be included. McKenna & Bell (1997) listed Nilopegamys and Colomys adjacent to one another, but I suspect that this is an admission of ignorance more than anything else. The fantastically-named Congo forest mouse Deomys ferrugineus (sometimes called the Link mouse) is another similar long-footed form of uncertain phylogenetic position.

You might be surprised to hear that there are amphibious mice at all. But not only are there several amphibious African mice, there are in fact multiple murid lineages around the world whose members swim, wade, or forage in aquatic environments. South America is home to an endemic amphibious murid radiation, Ichthyomyini, that consists of five genera: three whose members are generally termed fish-eating rats (Ichthyomys, Antomys and Neusticomys), as well as Rheomys (the Central American water mice) and Chibchanomys (the Chibchan water mice). Only distantly related to these are Holochilus (the web-footed rats), Nectomys (the Neotropical water rats) and the recently discovered, poorly known Lundomys and Amphinectomys, all of which seem to be part of the rice rat [oryzomyine] group. Then there’s Scapteromys (the ‘rata acuatica’), which seems to be closest to the deeply weird Kunsia and Bibymys (all three genera have been united by some workers in a little clade termed Scapteromyini). Australasia has an assemblage of amphibious murids that includes Hydromys (the beaver rats) and Crossomys (the earless water rat), plus a number of genera where experts disagree as to whether the animals are actually amphibious or not. And there are others elsewhere in the world.

What makes the African taxa special is that, not only have they so far failed to become as well adapted for amphibious life as have murids elsewhere (like the ichthyomyines, or Hydromys or Crossomys), but they might also be doing something that murids elsewhere are not. For, while Colomys, Malacomys and Deomys are even less specialized for amphibious life than Nilopegamys is, they are specialized in one, peculiar way: recall that, earlier on, I characterized them as ‘mice on stilts’. Noting that these genera possess particularly narrow, elongate feet, Kerbis Peterhans & Patterson (1995) showed that they formed a distinct cluster in terms of foot length : breadth, when compared with other murids. What might this mean? Unfortunately very very few published accounts discuss, describe or even mention the natural history and behaviour of these species, but a few do. While ichthyomyines and ichthyomyine-like murids are speedy swimmers that dart rapidly away underwater when threatened, Kerbis Peterhans & Patterson (1995) reported observations (made by Jonathan Kingdon and Ivan Sanderson) indicating that the stilt-legged mice really do use their long, narrow feet like stilts, wading around in shallow water. Such stilt-legged, wading mice seem unique to Africa.

Could it be that murids have evolved in this direction because there’s something unique about tropical African waterways that has allowed them to specialize in this way? One thing does spring to mind: the presence of amphibious shrews and otter-shrews, all of which are, also, uniquely African. Living alongside Colomys, Malacomys and Deomys are the shrews Ruwenzorisorex and Scutisorex, both of which reportedly exploit aquatic environments (though, to be honest, you wouldn’t know this from the literature). Murids are thought to have gotten into Africa relatively recently (about 6 million years ago), whereas the lipotyphlans have an African record going back as far as the Miocene.

Kerbis Peterhans & Patterson (1995) therefore suggested that ‘Prior or more successful exploitation of the ‘swimmer’ niche by lipotyphlans may have served to limit murid opportunities in this mode. Competition with lipotyphlans may also have driven the development of the ‘wader’ mode by central African forms’ (p. 346). Such amphibious lipotyphlans are entirely absent from South America, and this might then explain why ichthyomyines have radiated so extensively. But if this is true, what about Nilopegamys, which (as I said earlier) is an African form apparently evolving toward an ichthyomyine-like condition? Well, it inhabits the Ethiopian plateau (and is in fact one of about 30 mammal species endemic to this region (Yalden & Largen 1992)), where there are no amphibious lipotyphlans.

Oh well… back to the tyrannosaurs.

The illustration above is a John Gould painting of Hydromys, taken from here.

Refs - -

Kerbis Peterhans, J. C. & Patterson, B. D. 1995. The Ethiopian water mouse Nilopegamys Osgood, with comments on semi-aquatic adaptations in African Muridae. Zoological Journal of the Linnean Society 113, 329-349.

McKenna, M. C. & Bell, S. K. 1997. Classification of Mammals: Above the Species Level. Columbia University Press, New York.

Osgood, W. H. 1928. A new genus of aquatic rodents from Abyssinia. Field Museum of Natural History, Zoological Series 12 (15), 185-189.

Steppan, S. J., Adkins, R. M., Spinks, P. Q. & Hale, C. 2005. Multigene phylogeny of the Old World mice, Murinae, reveals distinct geographic lineages and the declining utility of mitochondrial genes compared to nuclear genes. Molecular Phylogenetics and Evolution 37, 370-388.

Voss, R. S. 1988. Systematics and ecology of ichthyomyine rodents (Muroidea): patterns of morphological evolution in a small adaptive radiation. Bulletin of the American Museum of Natural History 188, 259-493.

Yalden, D. W. & Largen, M. J. 1992. The endemic mammals of Ethiopia. Mammal Review 22, 115-150.