Main

Epoxy casts of the La Ferrassie I Neanderthal thumb and index-finger bones were scanned with a Minolta Vivid 900 laser digitizer to produce three-dimensional polygon mesh models that are accurate to within ±20 micrometres (Fig. 1a, b). We used these bone models to generate an articulated computer model with Maya Unlimited software. Movements were simulated by user-assigned ranges of motion at the centre of rotation of each joint with the aid of Maya animation tools.

Figure 1: The laser-scanned Neanderthal La Ferrassie I thumb and index finger.
figure 1

a, b, Dorsal views of the thumb (left) and index finger (right) in a, their neutral position, and b, the fully flexed position. In b, the two digits make tip-to-tip contact; the thumb carpometacarpal (bottom) joint is flexed by 10° from its neutral position, the thumb metacarpophalangeal (T-mcp) joint is flexed by 35°, the interphalangeal (Ip) joint is flexed by 15°, the index-finger metacarpophalangeal (I-mcp) joint is flexed by 45°, the proximal interphalangeal (Pip) joint is flexed by 65°, and the distal interphalangeal (Dip) joint is flexed by 15°.

A saddle-shaped metacarpal-1 base is a key feature for the production of precision grips (that is, the pads of the thumb and the fingers are opposed)5. A three-dimensional morphometric analysis indicates that some Neanderthal metacarpal-1 bases approach a condyloid shape because they lack a highly developed palmar beak4. The La Ferrassie I metacarpal-1 base has a moderately developed palmar beak that is at the extreme of the modern-human range of variation4, making it likely that the range of movements of its trapezial–metacarpal joint is similar to that of modern humans.

In fact, given the open configuration of the Neanderthal trapezial–metacarpal-1 joint, all Neanderthal thumbs were probably more mobile than that of modern humans. But as it has been suggested that Neanderthal thumb movements were restricted1,6,7, we minimized the mobility of the thumb by using the middle range of published modern-human flexion/ extension values5,8 and by eliminating metacarpal-1 pronation entirely.

The movement of the index finger is another important factor for producing precision grips. The asymmetry of the second metacarpal head (one half of the index-finger knuckle joint), combined with actions of muscle contraction and joint ligaments, causes the finger to turn towards the thumb in full flexion9. Although Neanderthal metacarpal-2 heads are slightly less asymmetrical than those of modern humans4, we were again cautious and restricted movement at the metacarpophalangeal joint to flexion/extension only. We assigned published modern-human flexion/extension values to the interphalangeal joints9 because these joints are functionally equivalent in Neanderthals and modern humans.

Even allowing for significantly limited joint movements relative to modern humans, flexion of the thumb and index finger results in tip-to-tip contact (Fig. 1b). Musgrave1 suggested that the Neanderthal's short thumb and first-finger proximal phalanges could have inhibited their precision of movement, but our results indicate that this is not the case. As there is no significant difference between Neanderthals and modern humans in the locations of their muscle and ligamentous attachments2, there remains no anatomical argument that precludes modern-human-like movement of the thumb and index finger in Neanderthals.

The demise of the Neanderthals cannot be attributed to any physical inability to use or manufacture Upper-Palaeolithic-like (Châtelperronian) tools, as the anatomical evidence presented here and the archaeological evidence10 both indicate that they were capable of manufacturing and handling such implements.