Knuckle-walking is a form of quadrupedal walking in which the forelimbs hold the fingers in a partially flexed posture that allows body weight to press down on the ground through the knuckles. In technical terms, knuckle-walking is locomotion with the manus (Latin for hand) distally flexed on contact with the substratum.
Anthropologists once thought that the common ancestor of chimpanzees and humans engaged in knuckle-walking, and humans evolved upright walking from knuckle-walking: a view thought to be supported by reanalysis of overlooked features on hominid fossils.
Since then, scientists discovered Ardipithecus ramidus, a human-like hominid descended from the common ancestor of chimpanzees and humans. Ar. ramidus engaged in upright walking, but not knuckle-walking. This leads scientists to conclude that chimpanzees evolved knuckle-walking after they split from humans 6 million years ago, and humans evolved upright walking without knuckle-walking.
Chimpanzees and gorillas engage in knuckle-walking. This form of hand-walking posture allows these tree climbers to use their hands for terrestrial locomotion while retaining long fingers for gripping and climbing. It may also allow small objects to be carried in the fingers while walking on all fours. This is the most common type of movement for gorillas, although they also practice bipedalism.
Their knuckle-walking involves flexing the tips of their fingers and carrying their body weight down on the dorsal surface of their middle phalanges. The outer fingers are held clear off the ground. The wrist is held in a stable, locked position during the support phase of knuckle-walking by means of strongly ﬂexed interphalangeal joints, and extended metacarpophalangeal joints. The palm as a result is positioned perpendicular to the ground and in-line with the forearm. The wrist and elbow are extended throughout the last period in which the knuckle-walker's hand carried body weight.
There are differences between knuckle-walking in chimpanzees and gorillas: juvenile chimpanzees engage in less knuckle-walking than juvenile gorillas. Another difference is that the hand bones of gorillas lack key features that were once thought to limit the extension of the wrist during knuckle-walking in chimpanzees. For example, the ridges and concavities features of the capitate and hamate bones have been interpreted to enhance stability of weight-bearing; on this basis, they have been used to identify knuckle-walking in fossils. These are found in all chimpanzees but in only two out of five gorillas. They are also less prominent when found in gorillas. They are however found in primates that do not knuckle-walk.
It has been suggested that chimpanzee knuckle-walking and gorilla knuckle-walking are biomechanically and posturally distinct. Gorillas use a form of knuckle-walking which is "columnar". In this forelimb posture, the hand and wrist joints are aligned in a relatively straight, neutral posture. In contrast, chimpanzees use an extended wrist posture. These differences underlie the different characteristics of their hand bones.
The difference has been attributed to the greater locomotion of chimpanzees in trees, compared to gorillas. The former frequently engage in both knuckle-walking and palm-walking branches. As a result, to preserve their balance in trees chimpanzees, like other primates in trees, often extended their wrists. This need has produced different wrist bone anatomy and, through this, a different form of knuckle-walking.
Knuckle-walking has been reported in some baboons. It has also been suggested that fossils attributed to Australopithecus anamensis and A. afarensis had specialized wrist morphology that was retained from an earlier knuckle-walking ancestor.
Giant anteaters and platypuses are also knuckle-walkers. Pangolins also sometimes walk on their knuckles. Another possible knuckle-walking taxon was the extinct chalicotheres, which looked something like a cross between a horse and a gorilla. The ground sloths may have also walked on their knuckles.
Knuckle-walking tends to evolve when the fingers of the forelimb are specialized for tasks other than locomotion on the ground. In the gorilla the fingers are used for the manipulation of food, and in chimpanzees for the manipulation of food and for climbing. In anteaters and pangolins the fingers have large claws for opening the mounds of social insects. Platypus fingers have webbing that extend past the fingers to aid in swimming, thus knuckle-walking is used to prevent stumbling.
It has been argued that knuckle-walking of chimpanzees and gorillas originally started from fist-walking as found in orangutans. African apes most likely diverged from ancestral arboreal apes (similar to orangutans) who were adapted to distribute their weight among tree branches and forest canopies. Adjustments made for terrestrial locomotion early on may have involved fist-walking, later evolving into knuckle-walking.
Evolution of knuckle-walking
There are competing hypotheses as to how knuckle-walking evolved as a form of locomotion, stemming from comparisons between African apes. High magnitudes of integration would indicate homoplasy of knuckle-walking in gorillas and chimpanzees, in which a trait is shared or similar between two species but is not derived from a common ancestor. However, results show that they are not characterized by such high magnitudes, which does not support independent evolution of knuckle-walking. Similarities between gorillas and chimpanzees have been suggested to support a common origin for knuckle-walking, such as manual pressure distribution when practicing this form of locomotion. On the other hand, their behavioral differences have been hypothesized to suggest convergent evolution, or homoplasy.
Another hypothesis proposes that African apes came from a bipedal ancestor, as there are no differences in hemoglobin between Pan and Homo, suggesting that their divergence occurred relatively recently. Examining protein sequence changes suggests that Gorilla diverged before the clade Homo-Pan, meaning that ancestral bipedalism would require parallel evolution of knuckle-walking in separate chimpanzee and gorilla radiations. The fact that chimpanzees practice both arboreal and knuckle-walking locomotion implies that knuckle-walking evolved from an arboreal ancestor as a solution for terrestrial travel, while still maintaining competent climbing skills.
It is important to note that not all features associated with knuckle-walking are identical to the beings who practice it, as it suggests possible developmental differences. For example, brachiation and suspension are almost certainly homologous between siamangs and gibbons, yet they differ substantially in the relative growth of their locomotor skeletons. Differences in carpal growth are not necessarily a consequence of their function, as it could be related to differences in body mass, growth, etc. It is important to keep this in mind when examining similarities and differences between African apes themselves, as well as knuckle-walkers and humans, when developing hypotheses on locomotive evolution.
One theory of the origins of human bipedality is that it evolved from a terrestrial knuckle-walking ancestor. This theory is opposed to the theory that such bipedalism arose from a more generalized arboreal ape ancestor. The terrestrial knuckle-walking theory argues that early hominin wrist and hand bones retain morphological evidence of early knuckle-walking. The argument is not that they were knuckle-walkers themselves but that it is an example of "phylogenetic 'lag'". "The retention of knuckle-walking morphology in the earliest hominids indicates that bipedalism evolved from an ancestor already adapted for terrestrial locomotion. ... Pre-bipedal locomotion is probably best characterized as a repertoire consisting of terrestrial knuckle-walking, arboreal climbing and occasional suspensory activities, not unlike that observed in chimpanzees today". See Vestigiality.
It has however been suggested that knuckle-walking evolved independently and separately in Pan and gorillas and so was not present in the human ancestors. This is supported by the evidence that gorillas and chimpanzees differ in their knuckle-walking related wrist anatomy and in the biomechanics of their knuckle-walking.
Kivell and Schmitt note "Features found in the hominin fossil record that have traditionally been associated with a broad definition of knuckle-walking are more likely reflecting the habitual Pan-like use of extended wrist postures that are particularly advantageous in an arboreal environment. This, in turn, suggests that human bipedality evolved from a more arboreal ancestor occupying a generalized locomotor and ecological niche common to all living apes".
Arguments for the independent evolution of knuckle-walking  have not gone without criticism, however. A more recent study of morphological integration in human and great ape wrists suggests that knuckle-walking did not evolve independently in gorillas and chimpanzees, which "places the emergence of hominins and the evolution of bipedalism in the context of a knuckle-walking background." 
Related forms of hand-walking
Quadrupedal primate walking can be done on the palms. This occurs in many primates when walking on all fours on tree branches. It is also the method used by human infants when crawling on their knees or engaged in a "bear-crawl" (in which the legs are fully extended and weight is taken by the ankles). A few older children and some adults retain the ability to walk quadrupedally, even after acquiring bipedalism. A BBC2 and NOVA in The Family That Walks On All Fours reported on the Ulas family in which five individuals grew up walking normally upon the palms of their hands and fully extended legs due to a recessive genetic mutation that causes a non-progressive congenital cerebellar ataxia that impairs the balance needed for bipedality. Not only did they walk on their palms of their hands but could do so holding objects in their fingers.
Primates can also walk on their fingers. In olive baboons, rhesus macaques, and patas monkeys such finger walking turns to palm walking when animals start to run. This has been suggested to spread the forces better across the wrist bones to protect them.
- Stokstad, Erik (24 March 2000). "Hominid ancestors may have knuckle walked". Science. 287 (5461): 2131–2132. doi:10.1126/science.287.5461.2131a.
- Richmond, BG; Begun, DR; Strait, DS (2001). "Origin of human bipedalism: The knuckle-walking hypothesis revisited" (PDF). American Journal of Physical Anthropology. Suppl 33: 70–105. PMID 11786992. doi:10.1002/ajpa.10019.
- Gibbons, A. (1 October 2009). "A New Kind of Ancestor: Ardipithecus Unveiled". Science. 326 (5949): 36–40. PMID 19797636. doi:10.1126/science.326_36.
- Wunderlich, RE; Jungers, WL. (2009). "Manual digital pressures during knuckle-walking in chimpanzees (Pan troglodytes)". Am. J. Phys. Anthropol. 139 (3): 394–403. PMID 19170201. doi:10.1002/ajpa.20994.
- Tuttle, RH; Basmajian, JV. (1974). "Electromyography of brachial muscles in Pan gorilla and hominoid evolution". Am. J. Phys. Anthropol. 41: 71–90. doi:10.1002/ajpa.1330410110.
- Kivell T.L. and Schmitt, D. (2009). "Independent evolution of knuckle-walking in African apes shows that humans did not evolve from a knuckle-walking ancestor.". Proc. Natl. Acad. Sci. USA. 106 (34): 14241–6. PMC . PMID 19667206. doi:10.1073/pnas.0901280106.
- Hausfater, G. (1975). "Knuckle walking by a baboon (Papio cynocephalus)". Am. J. Phys. Anthropol. 43 (2): 303–6. PMID 1180360. doi:10.1002/ajpa.1330430217.
- Richmond, BG; Strait, DS. (2000). "Evidence that humans evolved from a knuckle-walking ancestor". Nature. 404 (6776): 382–5. PMID 10746723. doi:10.1038/35006045.
- Orr, CM. (2005). "Knuckle-walking anteater: a convergence test of adaptation for purported knuckle-walking features of African Hominidae". Am. J. Phys. Anthropol. 128 (3): 639–58. PMID 15861420. doi:10.1002/ajpa.20192.
- Fish, FE; Frappell, PB; Baudinette, RV; MacFarlane, PM (2001). "Energetics of terrestrial locomotion of the platypus Ornithorhynchus anatinus" (PDF). The Journal of Experimental Biology. 204 (Pt 4): 797–803. PMID 11171362.
- Tassy P. (1978). "Chalicotherium: le ‘‘cheval-gorille’’". La Recherche. 87: 283–285.
- Tuttle, RH. (1969). "Knuckle-walking and the problem of human origins". Science. 166 (908): 953–61. PMID 5388380. doi:10.1126/science.166.3908.953.
- Tuttle, RH. (2005). "Knuckle-Walking and the evolution of hominoid hands". American Journal of Physical Anthropology. 26 (2): 171–206. doi:10.1002/ajpa.1330260207.
- Williams, S.A. (2010). "Morphological integration and the evolution of knuckle-walking". J. Hum. Evol. 58 (2): 432–40.
- Matarazzo, S.A. (2013). "Knuckle-Walking Signal in the Manual Phalanges and Metacarpals of the Great Apes". Dissertations. (755): vi–129.
- Edelstein, S.J. (1987). "An Alternative Paradigm for Hominoid Evolution". Hum. Evol. 2: 169–74. doi:10.1007/bf02436404.
- Richmond et. al,, B.G. (2001). "Origin of Human Bipedalism: The Knuckle-Walking Hypothesis Revisited". Yearbook of Physical Anthropology. 44: 70–105.
- Dainton, M; Macho, GA (1999). "Did knuckle walking evolve twice?". Journal of Human Evolution. 36 (2): 171–94. PMID 10068065. doi:10.1006/jhev.1998.0265.
- Williams, S.A. (2010). "Morphological integration and the evolution of knuckle-walking.". Journal of Human Evolution. 58: 432–440. PMID 20409573. doi:10.1016/j.jhevol.2010.03.005.
- Parker, Sue Taylor; Mitchell, Robert W.; Miles, H. Lyn (1999). The Mentalities of Gorillas and Orangutans: Comparative Perspectives. Cambridge University Press. ISBN 0-521-58027-7.
- Rose, M.D. (1973). "Quadrupedalism in primates.". Primates. 14 (4): 337–357. doi:10.1007/BF01731356.
- Patel, BA. (2009). "Not so fast: Speed effects on forelimb kinematics in cercopithecine monkeys and implications for digitigrade postures in primates". Am. J. Phys. Anthropol. 140 (1): 92–112. PMID 19294733. doi:10.1002/ajpa.21039.
- Hrdlicka, A. (1931). Children Who Run On All Fours, and Other Animal-like Behaviors in the Human Child. McGraw-Hill, New York. OCLC 653743.
- Humphrey, N., Keynes, R. & Skoyles, J. R. (2005). "Hand-walkers : five siblings who never stood up". Discussion Paper. Centre for Philosophy of Natural and Social Science, London, UK.