news and views

Palaeontology Smart-winged pterosaurs David M. Unwin

Why did ancient flying reptiles have so much processing-power in the back of their brain? To provide highly responsive flight control, is 100 YEARS AGO an answer to emerge from an innovative analysis of pterosaur skulls. At the recent meeting of the British Association at Southport I heard numerous rains, not surprisingly, are rarely showing that pterosaurs had a remarkably complaints (repetitions of those I have heard fossilized, leaving a large gap in our bird-like brain — for example, it had at not a few previous meetings) by the Bknowledge of the anatomy of most reduced olfactory lobes and large, laterally general public, members of the Association, extinct organisms. Fortunately, in some ver- displaced optic lobes. This suggests that, like on the too technical character of the papers tebrates — mammals, birds, dinosaurs and modern birds, pterosaurs were usually more read before it. These complaints referred to pterosaurs — the brain fits so tightly into interested in what they could see than what all the sections except, perhaps, those of the braincase that its external features are they could smell. The pterosaur brain seems anthropology, geography, and educational faithfully reflected by the contours of the to have been relatively small when scaled science. One overheard too often to be inner surface of the bones that enclose against body mass, however, with the brains pleasant such remarks as “I am interested it. Unfortunately, opportunities to recover of both Rhamphorhynchus and Anhanguera in zoology, but what is the good of coming these data from fossil material are infrequent plotting below the limits for extant birds. to listen to such a paper as this? I have no and often involve destructive techniques, Witmer et al. propose,convincingly,that this idea what the speaker is talking about” — thereby excluding many valuable specimens is primarily related to differing ancestries: the paper, in one specific instance, was from consideration. birds inherited their grey matter from cytological, and of great value undoubtedly; High-resolution X-ray computed relatively big-brained theropod dinosaurs6, and, “I have not gained much by becoming a tomography, which has proved extremely whereas pterosaurs inherited theirs from member of the Association; the papers are helpful elsewhere in palaeontology1, offers relatively small-brained archosaurs7. all over my head.” These complaints are the possibility of looking inside braincases The most striking results concern brain being made by well educated men and and generating a digital cast without damag- structures called floccular lobes and semi- women interested in science, but not versed ing the fossil. Witmer and colleagues2 have circular canals. Floccular lobes extend out- in its technicalities... The general public successfully applied this new technique to wards and backwards from the rear part of have really some cause for complaint that two kinds of pterosaur (see page 950 of this the brain and are exceptionally large in their subscription has been obtained from issue). These are a poorly understood group pterosaurs,while semi-circular canals encircle them on a misunderstanding. of flying reptiles that flourished during the the floccular lobes and are involved in From Nature 29 October 1903. Mesozoic (between 251 million and 65 balance. In living vertebrates the orientation million years ago) and which remain the sub- of the semi-circular canals, in particular the 50 YEARS AGO ject of controversy3. The new work clarifies lateral canal,relates directly to the ‘alert’posi- The death of Dr. E. P.Hubble from a heart several aspects of pterosaur neural anatomy, tion usually adopted by the head during loco- attack, on September 28 at the age of sixty- and prompts some startling new ideas motion and other activities. Exploiting this three, has robbed the world of astronomy of regarding their locomotion and behaviour. association, Witmer et al. show that, whereas its leading authority on nebulæ... On his Witmer et al. looked at rare, uncrushed the head posture of Rhamphorhynchus and return home in 1919 after the First World skulls of two specimens. One was of probably all basal pterosaurs was normally War, Hubble was appointed to the staff of Rhamphorhynchus, a long-tailed, crow-sized horizontal, in Anhanguera and most, if not the Mount Wilson Observatory. He soon creature from the Upper Jurassic (163–144 all,other derived forms,the head was directed established his position as a leading worker million years ago). The other was of sharply downwards at about 30ᑻ. in the field of the nebulæ. He showed that Anhanguera,a large,short-tailed form dating This is an elegant piece of diffuse nebulæ owe their illumination to to the Lower Cretaceous (144–97.5 million work. But explaining stimulation by radiation from hot stars. He years ago). In trade jargon, Rhampho- the difference then turned to the classification of the spiral rhynchus and Anhanguera are respectively nebulæ and showed conclusively that they ‘basal’ and ‘derived’ — loosely put, ‘primi- are stellar systems lying outside our galaxy. tive’and ‘advanced’. He established a scale of distances from a The new findings confirm study of the brightest stars and Cepheid earlier studies4,5 variables in the nebulæ. Working with Humason and the 100-in. telescope, he established the velocity–distance law from the red shifts of lines in the spectra; this increased the reliable distances to which astronomers could plumb the depths of space up to 250 million light-years. He studied the law of the red shifts in all its bearings, aiming always at cutting down the Figure 1 Ground truth? Pterosaur head orientations inferred by Witmer et al.2, and their number of conflicting interpretations, and he interpretation in terms of posture when on the ground. a, The horizontal alignment of the lateral kept always an open mind on the kinetic or semi-circular canal, indicated by the red line, is consistent with a crouching posture and forward- other explanations of the red shifts. directed head in basal pterosaurs, represented by Rhamphorhynchus. b, In derived forms such as From Nature 31 October 1953. Anhanguera, the reorientation of the canal can be interpreted in terms of an upright position and a downward-pointing head. (Pterosaurs redrawn from ref. 10 and not to scale.)

910 © 2003 Nature Publishing Group NATURE | VOL 425 | 30 OCTOBER 2003 | www.nature.com/nature news and views in head orientation is not easy. Maybe, within the membranes, enabling pterosaurs 2. Witmer, L. M., Chatterjee, S., Franzosa, J. & Rowe, T. Nature suggest Witmer et al., it relates to the large to build up a detailed map of the forces ex- 425, 950–953 (2003). cranial crests borne by many derived perienced by the wings from moment to 3. Buffetaut, E. & Mazin, J.-M. (eds) Evolution and Palaeobiology of Pterosaurs Geol. Soc. Lond. Spec. Publ. No. 217 (in the press). pterosaurs, including Anhanguera, which moment. Processing via the floccular lobes 4. Edinger, T. Am.J.Sci. 239, 665–682 (1941). could have affected skull aerodynamics could have allowed them to respond very 5. Hopson, J. A. in Biology of the Reptilia Vol. 9 Neurology A during flight and required some reposition- rapidly, through localized contraction or (eds Gans, C., Northcutt, R. G. & Ulinki, P.) 39–146 ing of the head. But this is inconsistent with relaxation of muscle fibres within the (Academic, New York, 1979). 6. Larsson, H. C. E., Sereno, P. C. & Wilson, J. A. J. Vert. Paleontol. the recent discovery of large cranial crests membrane and coordination with fore- and 20, 615–618 (2000). 8,9 in several basal pterosaurs , and their occa- hind-limb movement. Equipped with their 7. Bennett, S. C. Zool. J. Linn. Soc. 118, 261–308 (1996). sional absence in Anhanguera for instance. ‘smart’ wings, pterosaurs would have had 8. Carpenter, K., Unwin, D. M., Cloward, K. & Miles, C. in Alternatively, could head depression be excellent flight control. Despite their anti- Evolution and Palaeobiology of Pterosaurs Geol. Soc. Lond. Spec. Publ. No. 217 (eds Buffetaut, E. & Mazin, J.-M.) related to feeding? Anhanguera and other quity, they could even have outperformed (in the press). derived pterosaurs have been interpreted as modern birds and bats. I 9. Dalla Vecchia, F. M., Wild, R., Hopf, H. & Reitner, J. 10 aerial fish-catchers , a feeding style that David M. Unwin is in the Museum für Naturkunde, J. Vert. Paleontol. 22, 196–199 (2002). would have benefited from a downward- Humboldt Universität zu Berlin, Invalidenstrasse 43, 10.Wellnhofer, P. The Illustrated Encyclopedia of Pterosaurs 192 (Salamander, London, 1991). directed skull — especially as it may have Berlin 10115, Germany. 11.Henderson, D. & Unwin, D. M. J. Vert. Paleontol. 19, 50A (1999). permitted some stereoscopy, enabling accu- e-mail: [email protected] 12.Tischlinger, H. & Frey, E. Archaeopteryx 20, 1–20 (2002). rate judgement of distance to a moving 1. Rowe, T. et al. Nature 410, 539–540 (2001). 13.Lu, J. Mem. Fukui Pref. Dinosaur Mus. 1, 19–28 (2002). target. Again, however, there are inconsi- stencies. Many derived pterosaurs, such as the flamingo-like, filter-feeding form Ptero- Medical technology daustro,were not airborne fishers.But several basal forms were, as a specimen of Rham- phorhynchus with a fish in its belly eloquently Balancing the unbalanced testifies. Yet they still fished successfully with Frank Moss and John G. Milton their level heads. A more persuasive answer to this problem Elderly but healthy people are often seriously injured in falls. Exploiting lies on the ground (Fig. 1). Like their reptil- the phenomenon of stochastic resonance, biological physicists have ian ancestors, basal pterosaurs with their designed a shoe with a vibrating insole that helps maintain balance. relatively short arms were condemned to walk with the body and head in a near- o stay upright, the human body uses motion-analysis cameras. They applied horizontal position, aligned with the lateral exquisitely complicated and delicate the vibratory stimuli at random times, semi-circular canal. By contrast, functional Tfeedback control, operating through unknown to the subjects, and at strengths studies11 suggest that derived forms used the sensory and central nervous systems1. that were slightly below each subject’s their relatively long arms to prop themselves It does so in ways similar to a performer threshold of perception.Analysis of tracings upright. But because they still needed to balancing a pole on a finger2, by applying of the marker motion quantified the thresh- see in front of them as they walked, this minute forces and torques through muscles old at which sensory feedback mechanisms required some restructuring of the skull in the feet, ankles and knees. But these mus- kicked in to stabilize sway, enabling the and its posture, one consequence of which cles must receive signals that convey infor- authors to statistically characterize the sub- would have been reorientation of the semi- mation on how much force to apply through jects’ balance control4. All subjects showed circular canals. which muscles. This information is initially clear improvement with the introduction Attractive as they are, these ideas do not supplied to the central nervous system by of vibratory random fluctuations (‘noise’), address the extraordinarily large size of the neurons located in various joints and the with the elderly group showing the largest floccular lobes in pterosaurs. Witmer et al. soles of the feet. With ageing, the sensory benefit. Other studies indicate that patients suggest that this region of the brain may thresholds of these neurons increase. The whose balance is impaired by stroke or by have been responsible for coordination of information they transmit is therefore nerve damage associated with diabetes the head, eye and neck, permitting gaze- degraded,impairing balance. might also benefit5. The insoles have not stabilization during flight. Such an ability As they report in the Lancet 3, J. J. Collins yet been tested on subjects in motion, how- would have been useful for aerial hunters and his colleagues have successfully tested ever, nor have they actually been shown that relied primarily on sight. But not all an idea about how information from the to reduce falls, for example during walking pterosaurs had such a lifestyle, so this is not periphery of the sensory system might be or climbing stairs. an entirely satisfactory explanation. enhanced. They find that the application of How does the beneficial effect arise? The Far more convincing, in my view, is a random vibratory stimulation to the soles information content of subsensory stimuli Witmer and colleagues’ proposal that the of the feet can reduce ‘postural sway’, and is known to be increased by the addition of floccular lobes were responsible for process- improve balance in both young and elderly noise through stochastic resonance6,7. This ing large volumes of sensory data generated human subjects. Inexpensive, easily con- phenomenon is rooted in the physics of by the wing membranes. This is a plausible structed insoles, made of viscoelastic sili- systems with thresholds, such as sensory idea, because in other vertebrates the floccu- cone gel, generated the stimuli. Three neurons.Such systems transmit information lar lobes receive sensory inputs from skin mechanical vibrators, about the size of three by means of markers, or action potentials, and muscles. New, extraordinarily well-pre- stacked coins,were embedded in each insole, which are generated when a threshold is served pterosaur material from Germany12 one in the heel and two in the forefoot crossed.Noise added to a subthreshold stim- and China13 shows that the wing membranes regions.They were driven by random voltage ulus increases the threshold-crossing rate, were highly complex, containing structural generators, and the resulting vibrations thus improving the quality of the transmitted fibres, blood vessels and a fine network of propagated throughout the gel. information,and the optimal noise intensity muscles.These features would have given the Collins et al. measured postural sway by yields the maximum improvement. Much wings the ability to collect and transmit monitoring the position of reflective markers has been written on the subject (for instance, sensory information about local conditions attached to each subject’s shoulder using for a review focused on perception, see

NATURE | VOL 425 | 30 OCTOBER 2003 | www.nature.com/nature © 2003 Nature Publishing Group 911