I MENTIONED in a recent post that 75% of all of the types of mammals that have ever lived are now extinct. We have a tendency to think of extant organisms as all there is. We know dinosaurs once lived and are now extinct, but you’d be hard-pressed to get the average person to name one major extinct non-dinosaurian group. Yet we don’t have to look very far into our own past to find many of them.


An article published recently in Nature explored the routes of diversification taken by early mammals, but particularly caught my attention due to its excellent mammalian family tree.

Mammal Tree of Life
Reprinted by permission from Macmillan
Publishers Ltd: Nature, 450, 1011-1019, 2007.

Only those lineages extending to the far right of the diagram have survived to the modern day. In the smaller inset the thickness of the bands for each group illustrate how common it was in various time periods. You can see the multituberculates, notable for their unusual tooth anatomy, were very common until the early Paleocene. Historically these were the dominant group of mammals, existing in the largest numbers for the longest time–and now long extinct. Their contemporaries, the spalacotheroids and eutriconodonts, were successful in the early Cretaceous but were in decline by the end of that period. At the root of the tree the groups shown in blue are mammaliaforms–not true mammals yet, but very similar. This group first appeared in the Jurassic and did not survive the Cretaceous. The Cretaceous ended with a global extinction that eradicated the dinosaurs, and many stem-mammals suffered the same fate. Others lingered into the Paleocene only to become extinct, possibly outcompeted in the rapid radiation of the therians.

Since the mammaliaforms evolved from egg-laying therapsids and the monotremes still preserve egg-laying, it’s likely that the early mammals reproduced by laying eggs up until node 3 on the diagram. The skeletal structure of the multituberculates suggests that they bore immature live young similar to those of the marsupials. Since the monotremes produce milk, but secrete it through pores, it’s likely that nipples evolved at node 3 or sometime after. This may have happened as late as the divergence of the eutherians and marsupials, but probably came earlier. Some researchers think that milk secretion originally evolved as a means of hydrating the thin, leathery shells of basal mammals’ eggs, and was only secondarily used as a means of feeding hatchlings. With viviparity there would be a strong selective drive towards formation of a nipple to more efficiently feed the young. Evidence for this hypothesis will be found if it is discovered that monotremes secrete milk to coat their eggs.

The mammals have been furry since their beginning. The first fur preserved in the fossil record belongs to Castorocauda, a Jurassic mammaliaform with a resemblance to modern otters or beavers. This animal had a double-layered coat, with an undercoat and longer guard hairs. It is likely that fur evolved long before Castorocauda, and might pre-date the mammaliaforms. Unfortunately integument does not often leave traces in fossils, so most of the time we see no traces of fur, feathers, or scales.

We tend to think of extinct groups as more “primitive” than extant groups, but the mammals radiated rapidly and many niches occupied today were occupied by now-extinct forms in the past. The next diagram shows some of the niches available and how they have been filled at various times. The presence of a gliding mammaliaform in the Jurassic is especially interesting.

Mammal niches
Reprinted by permission from Macmillan
Publishers Ltd: Nature, 450, 1011-1019, 2007.

Currently the information is more complete for Jurassic (more ancient!) species than Cretaceous, but this will change with new fossil discoveries. In the past twenty years we have about doubled the number of Mesozoic mammal genera known, and characterization of these groups has vastly improved.

The tree of life is rather like a dense bush, on which we see the outer foliage, but there is much more beneath the surface if only we will look.

Luo, Z.-X. “Transformation and diversification in early mammal evolution.” Nature 2007, 450, 1011-1019.