Imaging our ancestors

  • Days of hunger compel Peking Man (Homo pekinensis) to hunt for his food deep in the forests of China 300 000 years ago. Stalking his prey, he hides behind a tree and waits for the right moment to move in for the kill.
  • Most of the superficial soft tissues have been applied to a cast of the Sima 5 Homo heidelbergensis skull, found in 1992 by a team led by Juan-Luis Arsuaga in the Sima de los Huesos (Pit of Bones) in Atapuerca, Spain. Later, skin is applied to the soft tissue and skull, and to complete the reconstruction, facial hair is added to reveal the face of the 350 000-year-old man.
  • It"â„¢s a brand new day for a juvenile female Australopithecus afarensis 3,4 million years ago. Her fossilised remains, unearthed in Ethiopia"â„¢s Dikika region, are from the same species as an adult skeleton found in 1974 that was nicknamed "Lucy".
  • And then there was one"¦ Homo sapiens contemplates his future as the dominant species on planet Earth.
  • The Fossil Human Reconstruction and Research Team at the American Museum of Natural History: GJ Sawyer (left) and Viktor Deak.
Date:30 April 2007 Tags:, ,

Take an anatomist, a sculptor, a palaeoanthropologist, a make-up artist, a forensic anthropologist, a painter and a draughtsman, and combine them all in one person. Then spark that individual with a vivid imagination, and add a burning obsession to create accurate images of human faces from the remote past – visages that no living human has ever seen. Put several such people together, and you have the Fossil Hominid Reconstruction and Research Team at the American Museum of Natural History, the artist-scientists who created a newly published book titled .

As humans, we have a self-awareness and curiosity about our own past that is unique among our fellow creatures. When Europeans and Americans first became aware of the existence of Egyptian mummies – the preserved bodies of individuals who lived some 4 000 years ago – they flocked to museums to gaze at the shrivelled, leathery faces of the pharaohs and pyramid-builders.

As we view them laid out in glass cases, we instantly imagine how they must have appeared in life, in the streets or palaces of Thebes. But while these remarkable mummies continue to fascinate, they are not really very ancient. They are essentially ourselves. If we want to visualise the faces of humans and near-humans who lived 100 000 years ago – even a million or five million years ago – our best window into deep time is through the combination of art and science known as forensic anatomical reconstruction.

Forensic anatomical reconstruction is widely used today in missing-person and homicide investigations, as well as in physical anthropology, for visualising the appearance of extinct human ancestors. Applied to beings that lived and died many thousands, or even millions, of years ago, the final result is like timetravelling to capture impossible portraits.

How do we go about conjuring up faces from the dawn of humanity? First, we obtain polyester or urethane resin cast reproductions of the best fossils from humankind’s prehistoric past, many of which have been discovered during the past century – and some of the most important only within the past few decades. Upon these casts of ancient fossil skulls and bones, we carefully rebuild the soft tissue of the face, comparing its conformation and depth of tissue to those of living humans and apes. Our techniques, meticulously applied, can provide a three-dimensional picture of what our ancient relatives and ancestors probably looked like.

The earliest known attempts to artistically put flesh on bones for an exhibit are the elaborate anatomical sculptural works of Ercole Lelli of Bologna, Italy (1702-66). Lelli assembled and posed both human and animal skeletons, and meticulously sculpted the musculature over them in wax. So renowned were his accomplishments that they have survived to this day, and are still exhibited at the Anatomical Museum in Bologna, Italy, where they continue to elicit gasps of admiration from visitors.

Many other artists and anatomists would follow, each contributing to and improving on the technique until it has become more science than art. In the late 19th and early 20th centuries, the American Museum of Natural History in New York City became a centre for artistic reconstructions of extinct animals. Under the leadership of palaeontologist Henry Fairfield Osborn, artists and sculptors were commissioned to create scientifically accurate images of living, breathing dinosaurs, woolly mammoths, and prehistoric humans.

One of the Museum’s pioneers in reconstructing the faces of early man was the talented James Howard McGregor, who taught anatomy at Columbia University. A gifted sculptor as well as an anatomist and physical anthropologist, McGregor developed techniques that became the forerunners of those still in use today. Between 1914 and 1919, he created a series of bronze portrait busts of Neanderthal, early Homo sapiens, Java Man, and the (now discredited) Piltdown Man.

Using new tools, modern materials, and the remarkable array of more recent fossil discoveries, we have continued in McGregor’s tradition – and so we have dedicated this book to him. Probably the best example of McGregor’s work is his bust of the Neanderthal male from La Chapelle-Aux- Saints, which has delighted and inspired generations of young visitors to the Museum, including ourselves.

Various techniques can be used to reconstruct the face of an unknown individual. In recent years, using statistically “average” measurements of soft tissue depths, forensic investigators have programmed computers to rebuild faces over skulls for homicide and missing persons investigations.

This method can quickly and efficiently work out a rough likeness, but has little value in recreating the faces of ancient human and pre-human ancestors. Fossil skulls are fragmentary at best, and there is only inferential reference data for comparison in reconstructing muscles and tissue.

By examining a modern human skull, a skilled physical anthropologist can usually determine its sex, possible “racial” affinities, age at death, nutritional status, disease, and various subtleties reflected in the facial bones, teeth, size and shape of jaw, and degree of muscular strength. But with ancient hominids, we usually cannot start with a complete skull as a base, and must first reconstruct one from available fragments – sometimes from several different individuals of the same genus and species. Only when we have a scientifically accurate reconstruction of the fossil skull can we attempt soft tissue restoration.

During the last century, artists, anatomists and palaeontologists developed two methods or “schools” of forensic reconstruction. The American school, pioneered in the 1940s by the anatomist Wilton M Krogman, devised a technique in which individual tissue depth markers are plotted over the facial part of the skull. These markers or points are then connected by a network of clay strips. This lattice of clay strips, connecting individual tissue depth markers or points, is then filled in or blended into a smooth skin surface.

The result is a mannequin or “generic” human face. Though practical for some purposes, the technique gives crude results because it is based on average or “standard” tissue depths. It ignores fine details about the skull’s individuality, geometry, muscular scar topography and overall morphology, and therefore masks many clues.

Such images are most useful in missing persons cases, where a more general description will draw a larger number of interested parties to make inquiries and view the model. We cannot use it, however, to accurately reconstruct faces of very pre- and ancient humans, which are different from modern skulls in many important features.

The other basic technique, known as the “Russian method”, was developed by the Russian anatomist and palaeontologist Mikhail Gerasimov (1907-70). His method requires greater, time-consuming precision, but allows much more fine-tuning. Individual soft tissue anatomy (muscle layers, glands and fibro-fatty tissue) is meticulously rebuilt in clay, using tissue depth measurements where applicable. Eighteen basic facial muscles are sculpted in place.

The technique is best described as a “dissection in reverse”. The reconstructor uses the skull not only as an armature, but as a detailed anatomical guide in the soft tissue placement. He or she is able to utilise the skull’s peculiarities, asymmetries, pathologies, sexual dimorphic changes and muscular attachments, allowing not merely a generic likeness, but a portrait of a particular individual.

Gerasimov’s Russian method recreates the nose by extending one line from the bony bridge and another from the floor of the nasal opening, and then rounding off their point of intersection at the tip. The nostrils are on average 1,67 times the width of the bony nasal opening in the skull. Corners of the mouth lie directly below the inner edges of the irises in the eyes, or between the lateral edges of the canine teeth, which are usually about the same width.

Our reconstruction technique is based on the Russian method. We carefully build up clay models of the missing facial musculature, glands, fibro-fatty tissues, depth measurements (where applicable), and skin from the skull outwards. Starting with the deepest individual musculature tissues, the face is slowly created, layer by layer, according to what the geometry, size and muscular landmarks of the skull tell us.

With the more recent skulls, we have better comparative data than did our predecessors.

Where McGregor’s measurements, for instance, were based on cadavers, the ones we use today are taken from living people using ultrasonic echo location and CT scans, and are therefore more reliable.

Of course, tissue depth measurements derived from living (extant) populations can be used only as a rough guide for extinct ones. Both McGregor and Gerasimov used anatomical technique as well as tissue depth measurements for their Neanderthals – the same combination of methods that we still use today.

Today, however, we have a wide range of sophisticated materials to help us create vividly lifelike heads. To begin, we use high quality, low-shrinkage silicone rubber to make a negative mould of the skull, lower jaw and any associated upper torso skeletal elements. From this mould we produce a positive cast in a fast-setting urethane plastic that faithfully reproduces every minute detail.

Of course, palaeoanthropologists rarely find complete fossil skulls, let alone full skeletons. At best we have to work with only parts of a skull, jaws, teeth, and some limb elements. If the skull or skeleton (as in the case of our Neanderthal reconstructed skeleton) is incomplete, then we must mould and cast parts from other similar specimens, then combine them to create a complete composite.

In the case of the reconstructed Neanderthal skeleton, we used the main body parts from two nearly complete individuals, representative parts from a third, and many fragments from several others. We incorporated all the recovered parts of a single adult male from the site of La Ferrassie, France. In assembling this composite, we produced the first complete Neanderthal skeleton ever seen.

Our reconstructed Neanderthal skeleton has not only extended our knowledge of what one of these individuals looked like, but enables us to better understand the bio-mechanical functions of the bony anatomy in walking and running. It further gives us a comparison between other fossil hominid and modern body proportions.

Faces, however, are our main concern, and it is the skull that dictates where and how the soft tissue anatomy is built up. The structure of the australopithecine facial skeleton suggests that the lips and nose were similar to those in living apes and not at all human in form, yet the small canines and rather human-like teeth suggest a face not quite fully pongid, but different from all known living apes.

As we progress, however, to more human-like forms such as Homo habilis or Homo rudolfensis, we are confronted with a combination or mosaic of australopithecine (ape-like) and human morphologies. So we must adjust our reconstructions according to what the bones are telling us to do.

Finally, as we move on to such true humans as Homo ergaster, Homo erectus and Neanderthal, we are at last able to draw more and more upon our knowledge of modern populations. Living humans possess a unique soft tissue anatomy in the middle or cheek region of the face called the superficial muscular aponeurotic system – a mesh of nerves, blood vessels and fibro-fatty tissue that connects the underlying facial musculature to the skin.

This mesh of tissue and fat fills out the face. If we sculpt only the muscles and glands on the skull of H ergaster, we can produce a face, but it will have a very gaunt or sunken look. When we add the superficial muscular aponeurotic system, and refer to modern tissue depth measurements as guides, the face fills out and appears much more lifelike.

It is here, therefore, that we are able to use an interface between what we know of modern human anatomy and apply it to a prehistoric skull with reasonably credible results.

The penultimate touches are the skin and glass eyes, which are positioned in the centre of the eye sockets. The mould is then filled with a urethane rubber tinted with a flesh colour, which of course is the artist’s best guess. Finally, more colour is jud