I will review the evolution of materials from three perspectives: the formation of biological composites in evolution, the phenomenon of layering in evolution, and the effects of evolving new structures on materials. Based on contemporary scientific findings, these are the most likely scientific mechanisms that account for the evolution of specialized materials and surfaces.
I. Biological composites
If you examine any segment of an animal body, you will find that it is composed of many different types of material. A human forearm, for example, is composed of fat layers, skeletal muscle, lamellar bone, and various other liquids and solids. This is a veritable kaleidoscope of tissue types, from a highly mineralized matrix (bone) to loosely-associated cell populations (fat layers). This diversity arises from the maintenance of two key mechanisms at the organ level: cell differentiation and tissue microenvironments. By contrast, biomechanical function acts to constrain the types of materials that can form through the forces experienced in functional contexts.
Tissue microenvironments. The aggregation of cells into tissues is also required for composite formation. More specifically, it is the co-existence of these cell populations in communities that allow for these composites to become well-integrated [4-6]. Cell populations residing in organs and other structures are organized in a manner analogous to an ecosystem, with energy exchange and substrate formation being prime candidates for the maintenance of homeostasis. Therefore, there is a hierarchical dependence leading from protein synthesis by individual cells to organismal context and back again that determine what materials will be formed in response to environmental and natural selective pressures.
Biomechanical function. While it seems counterintuitive, many structures that exist in nature may not be optimized for biomechanical function. Those that are “optimal” in terms of function tend to be highly specialized structures , and are also likely to be made from highly unique materials. Those structures that are less specialized in terms of function, or perform multiple functions simultaneously, should be made out of materials with a higher tribological diversity. Ultimately, these are testable predictions, but can also be inferred through examining the role of layering and the evolution of downstream  morphological structures.
In fields such as BioMEMS and bionanotechnology, a central issue is how very small functional devices can be fabricated. Generally, there are two methods for fabrication in very small devices: etching features into an existing surface, and deposition of materials onto an existing surface.
Natural Microfabrication. In systems that are driven by natural processes, microfabrication cannot be top-down as engineering techniques generally are. A bottom-up strategy involves a mechanism called directed self-assembly . In self-assembly, the configuration of a molecular system is biased by the available free energy, and the result is a highly-ordered system. The directed component comes into play in the form of "templates" or building blocks that guide further self-organization . This is similar to the idea of building blocks in genetic algorithms, which appear to play a role in allowing for evolvable genotypes in silico.
One link between surfaces and materials fabricated through natural means and contemporary biomimetic engineering is the characterization of superhydrophobic surfaces. Hydrophobicity occurs when water molecules are repelled from a membrane rather than being absorbed by it. This is a feature characterized in nature by the lotus leaf, and is a design consideration in a range of BioMEMS applications.
Superhydrophobic surfaces are created by "dual-scale roughness". As the name implies, the surface exhibits a rough topology at two distinct spatial scales . This further implies that hierarchical structures, with each component performing either a distinct or overlapping role, is required for evolving such surfaces. Surface stiffness is also a factor in guiding self-assembly , and may play the role of templates discussed in the previous section.
Response to Function. While natural selection generally acts upon the entire organism, there may also be selective processes acting at multiple spatial scales within morphological structures. The structure of the Gecko's foot is an example of this. Like the lotus leaf, the Gecko's foot uses a hierarchical structure of grooves and cilia to maintain a highly specialized function [13-15]. This type of specialization is seen across plant and animal species as a diversity of biological attachment mechanisms.
While the "optimality of structure" idea presented as an outcome related to the biomechanical function of biological composites, is best exemplified by these systems, the existence of tribological diversity requires a different set of examples. To do this, we must turn to the evolution of a generalized trait, preferably one that plays a role in the biomechanics of an organism.
III. Evolution of the Jaw
In the evolution of the jaw, it appears that one change gives rise to another. The jaw evolved rather early in vertebrate phylogeny, constituting a group called Gnathostomes  and includes a diverse collection of forms. In mammals, the jaw allows for the mandible to articulate relative to the upper portion of the head in concert with muscle activity allows for chewing and active manipulation at the mouth opening. The range of motion in a jaw is variable, from the very large aperture of the hippopotamus to the more constrained mouth opening of a Primate.
Diagram showing the route from a joint (human temporomandibular joint, left) that can potentially transfer and/or dampen forces with regard to "downstream" morphological structures (right).
A single evolutionary origin (monophyly). In the case of specialized and structurally specialized materials, such as specific instances of mineralized tissue, materials will arise once in a common ancestor and then be retained among members of a single clade. For example, vertebrates have tended to keep their skeletons made of either cartilage or hard bone. Major innovations in bone types tend to be restricted to specific clades. In cartilagenous fishes, the large-scale use of cartilage in the skeleton results in a single tissue type with the ability to endure a greater range of mechanical deformations. In mammals, hard bone comes in a number of forms (e.g. cortical, cancellous, and lamellar bone), each with their own physical properties and ability to endure and respond to various stresses and strains.
Multiple routes to refinement. The other option, observed in the origin of compounds such as chitin, is that a precursor material will evolve first and then be modified in various lineages to perform specific roles in a morphological system. For example, the basic chitin molecule may be adapted to a create a number of specialized materials in different lineages. Likewise, a basic set of molecules can be configured into different topologies through layering and/or other self-organized structures to produce highly-specialized materials. This is true of spider silk, which is used to build webs and other structures essential to their survival. Spider silk is a secreted protein that when spun into a silk strand has impressive mechanical properties. However, the performance characteristics of the resulting silk vary greatly in terms of both species of spider and gland of origin.
While the tribological profile of a given organisms is both determined by and a determiner of the evolution trajectory of a species, this can be approached from two perspectives. Reductionists will likely focus on the molecular building blocks of specific materials, answering questions such as: what are the signals that initiate the production of highly specific proteins and signaling molecules? Meanwhile, systems-level scientists will be more interested in the process of material synthesis during development, and answer questions such as: how is the self-assembly of highly specialized coordinated by generic processes? Both perspectives will be needed to fully appreciate the role specialized materials observed from across the diversity of the natural world and apply their artificial analogues to engineering technologies.
 Courtesy: Imiaos.
 Mahoney, T.S. et.al (2001). Cell adhesion regulates gene expression at translational checkpoints in human myeloid leukocytes. PNAS USA, 98(18), 10284–10289.
 Poole, C.A. (1997). Articular cartilage chondrons: form, function and failure. Journal of Anatomy, 191, 1-13.
 Jeanes, A.I. et.al (2010). Cellular Microenvironment Influences the Ability of Mammary Epithelia to Undergo Cell Cycle. PLoS One, 6(3), e18144.
 Burdick, J.A. and Vunjak-Novakovic, G. (2009). Engineered Microenvironments for Controlled Stem Cell Differentiation. Tissue Engineering Part A, 15(2), 205-219.
 McNeill-Alexander, R. (1988). Elastic Mechanisms in Animal Movement. Cambridge University Press, Cambridge, UK.
 "downstream" in this case means structures whose change in function have an indirect effect on other structures.
 Romano, F. and Sciortino, F. (2011). Colloidal self-assembly: Patchy from the bottom up. Nature Materials 10, 171–173.
 Wang, D. and Mohwald, H. (2004). Template-directed colloidal self-assembly – the route to ‘top-down’ nanochemical engineering. Journal of Materials Chemistry, 14, 459-468.
 Shirtcliffe, N.J. et.al (2004). Dual-scale roughness produces unusually water-repellent surfaces. Advanced Materials, 16, 1929-1932.
 Kima, T.W. and Bhushan, B. (2007). Effect of stiffness of multi-level hierarchical attachment system on adhesion enhancement. Ultramicroscopy, 107, 10-11, 902-912.
 Gorb, S. and Scherge, M. (2000). Biological microtribology: anisotropy in frictional forces of orthopteran attachment pads reflects the ultrastructure of a highly deformable material. Proceedings of the Royal Society of London B, 267, 1239-1244.
 Arzt, E., Gorb, S., and Spolenak, R. (2003). From micro to nano contacts in biological attachment devices. PNAS, 100(19), 10603-10606.
 I refer interested readers to a comprehensive online bibliography covering research on the biology and engineering of surface adhesion and attachment mechanisms (Gecko-centric) hosted by the Robotics and Intelligent Machines Lab at the University of California-Berkeley.
 Tree of Life, Gnathostomes.
 flexible and pliable in response to functional needs. For example, bone (e.g. mineralized tissue) can be both rigid and flexible, depending on how it is loaded and degree of hysteretic response. Material compliance is not directly related to plasticity, although there may be some interesting parallels between the two phenomena.