When those of us who study extant animals think of mammals, we usually think of animals with fur that nurse their young with milk. Yet neither of these characteristics is of much use to a paleontologist. How do students of fossils recognize the remains of mammals? And can we find any relationship between the characteristics available to paleontologists and those used by students of modern mammals?
Two characteristics of mammals that are at least sometimes preserved in the fossil record are (1) the mammalian middle ear contains a chain of three bones, the malleus, incus, and stapes; and (2) the lower jaw of mammals consists of a single bone. In the therapsids, immediate ancestors of mammals that dominated terrestrial habitats during the Permian, the middle ears contained just one bone, the stapes, and the lower jaw was made up of several bones.
The ear of modern mammals consists of an external pinna or flap of skin and cartilage (usually under some muscular control); a relatively small tympanic membrane that is sunk into a pit; an air-filled cavity called the middle ear that contains the three middle ear bones; and the inner ear, including a fluid-filled coil of bone called the cochlea. Sound impinging on the tympanic membrane causes the membrane to vibrate. The vibrations are picked up by the the most external of the middle ear bones, the malleus, which has a process (the manubrium) that touches the tympanum. Movement of the vibrating tympanum causes the manubrium and rest of the malleus to vibrate, which passes the vibrations (sound energy) to the incus, and from it to the stapes. That bone vibrates against a window to the inner ear. This sets up movement in a fluid in the cochlea, which is detected by hair-like structures and sent as nervous impulses to the brain. The arrangement of malleus, incus, and stapes actually serves as a sort of lever system that magnifies the vibrations received on the tympanum, increasing their amplitude across the chain. Vibrations are also amplified in a simple piston system because the tympanum is larger in area than the window to the cochlea.
The single bone that makes up the mammalian lower jaw is the dentary. At its posterior end is an articular (condyloid) process, which articulates with a bone called the squamosal in the upper jaw. The dentary also usually has an enlarged area at its posterior end, the coronoid process. This process provides attachment for muscles that participate in chewing.
In therapsids, and other vertebrates with jaws as well, the lower jaw is made up of a number of bones, including the dentary and a series of additional bones, which in therapsids are concentrated in the rear half of the jaw and are collectively sometimes called the "postdentary" bones. The jaw joint is between one of these, the articular (note -- this is different from the articular process of the dentary!) and the quadrate of the upper jaw. Within the therapsids, there is a clear trend over time for an increase in size of the dentary, including formation of a coronoid process, and reduction of the postdentary elements of the lower jaw (in fact, this trend may have taken place independently in several groups). As the dentary increased in size, it eventually contacted the squamosal of the upper jaw. In some late therapsids and the earliest mammals, the jaw joint is actually double -- with contact both between articular and quadrate (the old way) and between dentary and squamosal (the new).
Even when they were a functional jaw articulation, the articular and quadrate may have passed sound vibrations received by the lower jaw to the stapes. In some therapsids, the angular, a bone located next to the articular, has a "reflected lamina," or bent plate, that looks like it might have served to support a membrane (tympanum) to detect vibrations. These vibrations were picked up by an arm of the articular bone that touched the tympanum (the retroarticular process), then passed through the jaw joint to the quadrate of the upper jaw, and from there, to the stapes. As the dentary increased in size, the articular and quadrate became smaller and less tightly attached to lower and upper jaw, respectively. As a result, these two bones were better able to transmit sound. The articular is known as the malleus in mammals, and it contacts the mammalian tympanic membrane its manubrium (the old retroarticular process). The reflected lamina of the angular forms the ectotympanic bone of mammals, a bony ring that supports the tympanic membrane. The therapsid quadrate becomes the incus, which still sits between malleus (articular) and stapes and transmits vibrations between them. As was certainly the case in the case of therapsids, movement of the stapes transmits sound to the inner ear, where it is turned into nervous impulses and sent to the brain.
Exactly how therapsid and early mammal ears functioned is not clear. A remaining problem concerns exactly how the transition from a tympanum in contact with the stapes to a differently-oriented tympanum contacting the malleus took place. One authority has suggested that for a time, there may in fact have been two tympanic membranes, the old one located towards the back of the skull and touching the stapes, and a new one located more anteriorally and on the outside surface of the lower jaw, touching the reflected lamina of the articular (Allin, 1986). There is little doubt, however, that such a change took place; not only is it clear from the fossil record, but the sequence of changes from an articular-quadrate joint to a dentary squamosal joint with the articular and quadrate participating in the middle ear can actually be seen in the developing young of opossums.
Thus, the development of the middle ear of mammals and the their single-bone lower jaw are part of the same package. Why might these changes have occurred? We don't know, but we can speculate. And what follows is indeed speculation; we probably cannot know the answer to this question.
First, the changes in the lower jaw are perhaps not surprising in a group under selection for increased efficiency of biting and chewing. The dentary bears the teeth. Sutures between bones are weak points; therefore, enlargement of the tooth-bearing bone, so that most of the forces resulting from chewing acts upon it alone, makes sense. Similarly, the growth of a coronoid process provided extra space for attaching muscle, especially muscle oriented to increase the strength of the bite at the point where upper and lower teeth meet and reduce the forces operating at the jaw joint itself (e.g., Carroll, 1988, p. 195; Crompton and Hylander, 1986).
Similarly, selection for improvements in hearing sensitivity might have resulted in less-firmly attached quadrate and articular, because this would have enabled them to vibrate more freely, transmitting sound energy from the angular and articular through the quadrate to the stapes.
Changes in the jaw and middle ear may have in a sense been driven by selection for endothermy. Generating the heat to maintain a high body temperature is very expensive; around 70% of the calories consumed by a cotton rat, for example, are used for this purpose (Randolph et al., 1977). This is a cost not paid by ectothermic animals, and assuming it places a premium on efficient use of resources. Endothermy may also be correlated with activity at night, i.e., under conditions when reliance on hearing, smell, and touch may have replaced reliance on sight. The evolution of other quintessential mammalian traits may have been related: a zygomatic arch, formed as the temporal fenestra enlarged to accomodate an increasingly large temporal muscle; a secondary palate to separate the passage of air and food through the mouth; teeth differentiated along the jaw (and often highly specialized); changes in the limbs tending to bring them (especially hind limbs) under the body, increasing efficiency and speed of locomotion; hair for a sense of touch and for insulation; lactation to permit rapid growth (small young, because of their high surface area to mass ratio, lose heat rapidly and have difficulty maintaining high body temperature; they therefore need a rich and concentrated food source to allow them to keep warm and grow rapidly); and diphyodonty (itself made possible by lactation) so that precise occlusion can be maintained.
Jaws with a single bone, middle ears with three bones, lactation, hair, etc. are clearly not the only solution to the problem of providing enough energy to maintain a homeothermic, endothermic lifestyle. Birds and dinosaurs went down a different path. But it is what mammals appear to have done, and it seems to have worked.
Phil Myers (author).
Allin, E. F. 1986. The auditory apparatus of advanced mammal-like reptiles and early mammals. Pp. 283-294 in N. Hotton III, P. D. MacLean, J. J. Roth, and E. C. Roth, eds. The Ecology and Biology of Mammal-like Reptiles. Smithsonian Institution Press, Washington. x+326 pp.
Crompton, A. W., and W. L. Hylander. 1986. Changes in mandibular function following the acquisition of a dentary-squamosal jaw articulation. Pp. 263-282 in N. Hotton III, P. D. MacLean, J. J. Roth, and E. C. Roth, eds. The Ecology and Biology of Mammal-like Reptiles. Smithsonian Institution Press, Washington. x+326 pp.
Jenkins, F. A., Jr. 1984. A survey of mammalian origins. Pp. 32-47 in P. D. Gingerich and C. E. Badgley, eds. Mammals. Notes for a Short Course. University of Tennessee Dept. of Geological Sciences Studies in Geology #8. iv+234 pp.
Randolph, P. A., T. C. Randolph, K. Mattingly, and M. M. Foster. 1977. Energy costs of reproduction in the cotton rat, Sigmodon hispidus. Ecology, 58:31-45.