Magazine R61 Conclusion the fine arts [7]. To see if this also Questions of scale in biology have Correspondence applies to museums, veterinary a rich history, and an exciting books and toy shops, we collected future. The investigation of how hundreds of walking depictions and life copes with changes in size Erroneous tested whether or not they correctly has unquestionably advanced our display limb positions. We found understanding of basic biology. quadruped walking that almost half of the depictions Nanotechnology, microfabrication, depictions in natural are wrong. This high error rate in and microelectronics are providing walking illustrations in natural history new tools for biological investigation. history museums museums and veterinary anatomy They make it possible to sense and books is particularly unexpected in a perturb previously inaccessible Gábor Horváth1,*, Adelinda Csapó1, time where high-speed cameras and microscopic life in more and more Annamária Nyeste1, Balázs Gerics2, the internet offer ideal possibilities to sophisticated ways. Less appreciated Gábor Csorba3 and György Kriska4 obtain reliable quantitative information but equally important is that for about tetrapod walking. larger organisms they enable sensing Since the work of the photographer Although humans have observed and perturbation of multiple parts Eadweard Muybridge in the walking quadrupeds for thousands of of intact, freely behaving animals, 1880s [1,2], experts know well years, the exact characterization of in complex or even native habitats. how quadruped animals walk. All the walking of tetrapods had to wait As we move towards integrated walking tetrapods advance their for the advent of photography [1,2]. measurement of metabolism, legs in the same sequence, and The usual sequence by which the biomechanics, and neural control in only the timing of supporting feet legs of walking quadrupeds contact freely behaving animals, the future for may differ [3–6]. Given the long time the ground, the so-called ‘foot-fall questions of scale in biology looks since Muybridge’s work, one would formula’, is: left hind leg–left foreleg– extremely bright. assume that this knowledge should right hind leg–right foreleg (LH–LF– be reflected in the depictions of RH–RF). The biophysical reason for Acknowledgments walking quadrupeds made by work this uniformity is that this gait confers The author would like to acknowledge of painters, taxidermists, anatomists maximal static stability to the body [6]. the generous and insightful comments and toy designers. The postures of To study how correctly this foot- received from Mark Frye, John Hutchinson, legs of walking horses, however, are fall formula is represented in natural Jim Usherwood, and Michael Dickinson, in frequently erroneously illustrated in history museums, veterinary books addition to the members of the Structure & Motion Laboratory. Further reading A B Alexander, R.M. (2003). Principles of Animal Locomotion. (Princeton, N.J.: Princeton University Press.) Biewener, A.A. (1989). Mammalian terrestrial locomotion and size. Biosci. 39, 776–783. Biewener, A.A. (2005). Biomechanical consequences of scaling. J. Exp. Biol. 208, 1665–1676. Feder, M.E. (1987). New Directions in Ecological Physiology. (Cambridge: Cambridge LH LF University Press.) RH Glazier, D.S. (2008). Effects of metabolic level RF on the body size scaling of metabolic rate in birds and mammals. Proc. R. Soc. Lond. B 275, 1405–1410. McMahon, T.A., and Bonner, J.T. (1983). On Size C D and Life. (New York: Scientific American Library.) Orlovski, G.N., Deliagina, T.G., and Grillner, S. (1999). Neuronal Control of Locomotion: From Mollusc to Man. (Oxford: Oxford University Press.) Peattie, A.M., and Full, R.J. (2007). Phylogenetic analysis of the scaling of wet and dry biological fibrillar adhesives. Proc. Natl. LF Acad. Sci. USA 104, 18595–18600. RF LF LH RH LH Schmidt-Nielsen, K. (1984). Scaling: Why Is RH Animal Size So Important? (Cambridge: RF Cambridge University Press.) Szirtes, T., and Rózsa, P. (2007). Applied Current Biology Dimensional Analysis and Modeling, 2nd edn. (Amsterdam; New York: Elsevier/Butterworth- Heinemann.) Figure 1. Erroneous three-foot-supported walking depiction of an aardwolf (Proteles cristatus). Structure and Motion Laboratory, Royal (A) Sample at the Natural History Museum, Florence, Italy (photo by Balázs Gerics) and its leg pos- Veterinary College, Hawkshead Lane, ture (B). (C,D) Two possible corrections. Erroneously, stepping by the right hind leg is followed by Hatfield, Hertfordshire AL9 7TA, UK. raising the left foreleg, which does not occur in quadruped walking. Instead, it should be followed by E-mail: [email protected] raising right foreleg (C), or raising left foreleg should be preceded by the step of left hind leg (D). Current Biology Vol 19 No 2 R62 Since the 1880s, knowledge of A B correct representations of quadruped walking is available from the publications of Muybridge [1,2] and others [3–7]. Our assumption, that the majority of the walking depictions may RH be correct, turned out to be wrong: LH LF RF 41.1–63.6% (on average 46.6%) of them are erroneous. Thus, there is almost 50% chance to come across an incorrect walking depiction in museums, anatomy books [8–10], or toy shops. Hence, taxidermists, C D book illustrators and toy designers are nowadays still not completely aware of the quadruped walking, despite the fact that numerous scientific tools are available to study RH the animal motion quantitatively, and LH RF LF LH RH LF RF to circulate the gathered information among communities concerned. As we show here, there are many erroneous depictions of quadruped walking even in the scientific world, and these errors Current Biology can even be propagated given the ease of modern information exchange. Figure 2. Erroneous three-foot-supported depiction of a domestic dog (Canis familiaris). (A) Display at the Natural History Museum, Oulu, Finland (photo by Gábor Horváth) and its leg pos- Supplemental data ture (B). (C,D) Two possible corrections. Erroneously, stepping by left hind leg is followed by raising Supplemental data are available at http:// right foreleg, which does not occur during walking. Instead, it should be followed by the step of left www.current-biology.com/supplemental/ foreleg (C), or raising right foreleg has to be preceded by the step of right hind leg (D). S0960-9822(08)01633-3. References and quadruped toys, we gathered foot-supported depictions; rtotal = 1. Muybridge, E. (1881). Attitudes of Animals in numerous walking depictions from 46.6% for the total 307 walking Motion (California: Stanford University Press). various sources and analysed them illustrations. 2. Muybridge, E. (1887). Animal Locomotion with respect to the foot-fall formula. Considering only the two- and (Philadelphia: Pennsylvania University Press). 3. Hildebrand, M. (1965). Symmetrical gaits of The postures of the fore- and hindfeet three-foot supported illustrations of horses. Science 150, 701–708. of these depictions were compared horses, or related quadrupeds (zebra, 4. Gambaryan, P.P. (1974). How Mammals Run: Anatomical Adaptations (New York: John Wiley with the corresponding real positions donkey, deer, elk, antelope, muntjac, and Sons). of supporting and lifted feet for the kudu, dik-dik, impala, gazella, bongo, 5. Jayes, A.S., and Alexander, R.M. (1978). eight typical stride phases of walking duiker, nyala, oribi, okapi), or both Mechanics of locomotion of dogs (Canis familiaris) and sheep (Ovis aries). J. Zool. horses (see Supplemental Data horses and related tetrapods, we (London) 185, 289–308. 6. Hildebrand, M. (1989). The quadrupedal gaits published with this article online). We obtained: rhorse = 50.4%, rhorserelated = of vertebrates. Bioscience 39, 766–775. studied only illustrations in which the 43.4%, rhorse+horserelated = 48.2%. 7. Brown, L. (1968). The importance of proper animals were on horizontal substrates Hence, the error rate for horses and tripods. Curator 11, 7–32. and lifted one or two legs. Distinction related quadrupeds is about the same 8. Fehér, G. (1980). Functional Anatomy of Domestic Animals. vol. I. Organs of Motion, of walking depictions from illustrations as that for the total 307 depictions (Budapest: Mezőgazdasági Kiadó, in of other gaits/behaviours was made studied. Not surprisingly, the error Hungarian). 9. Callegari, E. (2003). Museo di Anatomia degli on the basis of leg postures and the rate rmuseum = 41.1% is very similar to Animali Domestici. Edagricole (in Italian). attitudes of trunk, head, neck, mane, rtaxidermy = 43.1%, because taxidermy 10. Szunyoghy, A., and Fehér, G. (2004). Small tail and hair. In total, we analysed companies provide museums with drawing school, artistical comparative anatomy, (Budapest: Novella Kiadó, in 307 two- and three-foot supported quadruped models. The small Hungarian). depictions, which were collected difference between rmuseum = 1Eötvös University, Physical Institute, randomly and representatively. Figures 41.1% and rtoy = 50%, and in 1 and 2 show examples of incorrect particular the relation r = Department of Biological Physics, H-1117 toy Budapest, Pázmány sétány 1, Hungary. walking depictions from museums. 50%<rbook = 63.6% are, however, 2Szent István University, Faculty of Veterinary The error rates (r) of the investigated unexpected, because the quadruped Science, Department of Anatomy and depictions were: rmuseum = 41.1% in toy models are intended for children Histology, H-1078 Budapest, István u. 2, 3 natural history