336 NOTES AND COMMENTS

Transcription

1 336 NOTES AND COMMENTS RILLING, M Stimulus control and inhibitory processes, pp In W. K. Honig and J. E. R. Staddon (eds.), Handbook ofoperant Behavior. Century. N.Y., USA. RYAN. M. J., J. M. Fox. W. WILCZYNSKI, AND A. J. RAND Sexual selection for sensory exploitation in the frog Physalaemus pustulous. Nature 343: SPENCE, K. W The differential response in animals to stimuli varying within a single dimension. Psych. Rev. 44: SUBOSKI, M. D Releaser-induced recognition learning. Psych. Rev. 97: TEN CATE, C, AND P. BATESON Sexual selec- tion: The evolution of conspicuous characteristics in birds by means ofimprinting. Evolution 42: VON SCHANTZ, T., G. GORANSSON, G. ANDERSSON, I. FROBERG, M. GRAHN, A. HELGEE, AND H. WITTZELL Female choice selects for a viability-based male trait in pheasants. Nature 337: WEISS, S. J., AND R. D. WEISSMAN Generalization peak shift for autoshaped and operant key pecks. J. Exp. Anal. Behav. 57: ZAHAVI, A Mate selection: A selection for a handicap. J. Theor. BioI. 53: Corresponding Editor: R. Thornhill Evolution. 47(1) pp FLIGHT CAPABILITIES IN ARCHAEOPTERYX J. R. SPEAKMAN Department 0/ Zoology. University ofaberdeen, Aberdeen. AB9 2TN Scotland. UK Key words.-s-archaeoptervx. birds, energetics. efficiency, flight, muscle power, origins. Received June 13, Accepted April Archaeopteryx lithographica lived in the late Jurassic period in the area of modern day Bavaria. Germany (Wellnhofer, 1990). It has aroused considerable interest because it has a mixture of features. some unmistakably avian, i.e., feathers (Ostrom, 1985), and others. notably in the skeleton. unequivocally reptilian (Ruben, 1991). As such, Archaeopteryx represents an important stage in the evolution of powered flight by the birds (Ostrom, 1974, 1976). Much speculation has been directed to evaluating the exact flight capabilities ofarchaeopteryx. Until recently it had been suggested that Archaeopteryx was terrestrial and cursorial, but ranged into the arboreal habitat (Wellnhofer, 1990). It was thought Archaeopteryx could glide and make feeble flapping flights down from trees. but it was incapable of taking off directly from the ground (Dodson, 1985) even after leaping into the air with a running start (Rayner, 1985). This interpretation rests primarily on inferences relating to the amount ofmuscle that Archaeopteryx had available for generatingthe powerrequirementsfor flapping flight. Its pectoral muscles and deltoids are together thought to have averaged only about 9% of the body mass (7% in the pectorals: Ruben, 1991), which is considerably less than the average of 25% of body mass found in modern birds. Thus it has been suggested that sustained flapping flight would be impossible. Although the range of flight muscle sizes in modern birds is large (11.5 to 43.9%: Marden, 1987), there is a marginal muscle ratio of 16%, below which it has been suggested that birds are incapable of ground upward take-off from a stationary position. Only II out of 425 modern birds examined had flight muscles comprising less than 16% of their body mass (Marden, 1987) and all these were eithergrebes (Podicipedidae) which are weak fliers that cannot take offwithout a long running start, or secretive rails (Rallidae) whose flight behaviors are unknown and which are possibly flightless (Marden, 1987). Since Archaeopteryx had muscles (comprising 9% of body mass) which fall well below the marginal limit, the implication was that not only would it be incapable of sustained flight but it would also be incapable ofground upward takeoff. Flight would consequently have had to evolve 'from the trees down' (Dodson. 1985). It is difficult to gauge the accuracy of the estimates of body mass and in particular the flight muscle mass for Archaeopteryx. since both have been inferred by necessity from examination and extrapolation from a relatively small sample of animals (see debate on estimated mass in Yalden, 1971). The estimate of flight muscle mass is inferred from the areas of muscle insertion, and since the anatomy ofarchaeopteryx differs significantly from that ofmodern birds (Ostrom, 1976) it is possible only to suggest that the area available for insertion is significantly less than that found in modern birds. Despite these potential problems in the accuracy of the estimate of the body mass and flight muscle masses ofarchaeopteryx, it has recently been suggested that even if the estimated 9% of body mass for the flight musculature is correct, this does not necessarily mean Archaeopteryx was incapable of sustained flapping flight, nor even ofground-upward takeofffrom a standstill (Ruben, 1991). The reasoning behind this (r)evolutionary suggestion is that previous assessments of the flight capabilities of Archaeopteryx have assumed that the muscles during burst activity are performing in a fashion characteristic ofliving birds. However, Ruben (1991) suggested that reptilian muscle is capable of generating at least twice the avian power output per unit mass. Consequently, although Archaeopteryx had relatively small muscles these need not necessarily have limited its flight capabilities if these 1993 The Society for the Study of Evolution. All rights reserved.

2 NOTES AND COMMENTS 337 Power (Watts) FIgure lb 100 Effloency ('"to) 50. Ma)umum aerobe runnmg speed Total costs cncccmouon (AerobiC plus Anaerobe).. Melabol,c Input Running Speed Running Speed Burst Speed ;-=.....,. Ac1uaf efficiency FIG. I. (panel a) Metabolic energy input and mechanical power output as functions of running speed for reptiles. At low speeds the metabolic input is met entirely aerobically. At point A the animal reaches its aerobic scope. Over these speeds the efficiency (panel b) measured as either mechanical output/aerobic expenditure or mechanical output/total expenditure is low and constant. At greater running speeds than the maximum aerobic speed an increasing portion of the metabolic costs are met by anerobis. Total metabolic input and mechanical output continue to increase as speed increases above the maximum aerobic speed hence actual efficiency [mechanical output/total metabolic input (C in la)] remains low (panel b-dotted line). As the mechanical output and aerobic contribution lines converge the apparent efficiency [mechanical output/aerobic metabolic input (B in Ia)] increases exponentially and may reach or exceed 100%. muscles were functionally reptilian rather than avian. With 7% of its approximately 200 g body mass composed of pectoralis flight muscle. and this muscle capableofgenerating450w.kg I (versus about 190W.kg I for typical avian muscle). the power available to support flight would be 6.3 W (Ruben Fig. 6). This available power. from the pectorals. is at least twice at great as the power requirements for sustained flapping flight calculated from the aerodynamic model of Pennycuick (1989). Furthermore. with muscle that is functionally twice as capable as avian muscle Archaeopteryx would have had a total flight muscle complement (pectorals and deltoids) equivalent to 18% of the body mass (i.e.. 9% x 2). This would be greater than the marginal muscle volume of 16% (Marden. 1987) and hence sufficient to enable the takeoff from a standstill at an angle of 30 (Ruben Fig. 7). In this paper the argument developed by Ruben (1991) will be challenged from two perspectives. First. the balance between available and required power will be considered, and second the calculated capacities for power output by avian muscle will be addressed. In Ruben's(l991, Fig. 6) comparison ofthe available power from the flight muscle of Archaeopteryx compared to the aerodynamic requirements. it is clear that Archaeopteryx is easily capable of powered flight since its available power is more than twice that required. To evaluate the available power Ruben calculated the metabolic power capacity of muscle from studies of oxygen consumption during locomotion. combined with the measured sprintspeeds and muscles masses of running reptiles. This available power was then compared with the power requirements for flight [i.e.. the aerodynamic power requirements based on a mathematical flight cost model (Pennycuick. 1989)]. However, this latter model considers only the mechanical power output and this grossly underestimates the actual metabolic power requirements for a flying animal because muscle is not 100% efficient at generating mechanical power (Stevenson and Josephson, 1990). To evaluate the metabolic costs offlight the mechanical power costs from the model must be multiplied by an estimate of muscle efficiency, or alternatively metabolic power inputs for sustained flight can be measured directly. A more recent review of flight energy requirements based on the metabolic power inputs of flying birds and bats (Speakman and Racey. 1991) reveals that although the mechanical power output requirements for a 200 g Archaeopteryx with a 0.6 m wingspan. based on the Pennycuick model. average around 3 W (Ruben. 1991). the actual metabolic power a 200 g animal would have to providefrom its muscles would be in the region of 16.6 Watts (prediction based on the allometric equation in Figure 3. Speakman and Racey. 1991). This latter estimate exceeds by far the calculated available power from the pectoral muscles (6.3 W. SE = 0.64 from the coefficient of variation in estimated muscle power capacity). even assuming Archaeopteryx had the enhanced power generating abilities of reptilian muscle. If I add a further 2% of body mass to the muscle mass to account for the deltoids. which are also involved in the flight power generation (Ruben. 1991). and subtract from the total required energy expenditure in flight an (avian) estimate of BMR to account for energy expenditure not involved in flight. the available power still fails by far to meet the estimated requirements. On this basis I suggest Archaeopteryx would clearly be incapable of sustained powered flight. Ruben (pers. comm.) has suggested the use of an assumed 100% efficiency for muscle contraction is justified on the basis of observations that as animals increasingly support their activity by anerobis the calculated efficiency (mechanical power output/aerobic metabolic power input) increases exponentially and approaches or even exceeds 100% (see for example Smit et al., 1971). This argument. however. is based on a misinterpretation of the calculated efficiency. At slow walking speeds reptiles are aerobic. The relationship between oxygen consumption (power input) and speed is linear. Since mechanical power output is a function of body mass multiplied by velocity the mechanical power output also increases linearly but the mechanical output curve is much lower than the metabolic input

3 338 NOTES AND COMMENTS TABLE I. Calculated metabolic power inputs ofavian and reptile muscle during terrestrial locomotion. For all animals the power is calculated after the method used by Ruben (1991). That is the net cost oftransport (oxygen consumption) is converted to energy assuming an oxy-calorific equivalence of 20.9 Joules per ml 02. This is then multiplied by the estimated maximum sprint speed to obtain the mass specific maximum power (W.kg I body mass), and this latter value is multiplied by the reciprocal of the proportion of the total body mass that comprises locomotory muscle (see also Ruben, 1991, Fig. 4). NCT - Net cost of transport [slope ofrelationship between mass specific oxygen consumption and locomotion speed (ml, 02/kg- 11m-I) for birds from Fedak and Seeherman (1979) and for reptiles from John-Alder et al. (1986)], BM-Body mass (kg). Maximum sprint speed (m/s 1) [for birds predicted from allometricequationofmammalian maximum sustainableaerobic speeds against mass, Garland (1983) and for reptiles from estimates ofburst speeds at body temperatures of 35-40"C (Bennett, 1980; Garland, 1984, 1985; John-Alder et al. 1986), Power (W.kg- 1 body mass) and MS Power-Muscle specific power output (W.kg- 1 muscle mass). The proportion of body mass accounted for by the locomotory muscles was estimated from Hartman (1961). In the absence ofcomplete data I assumed the estimate for quail (Coturnix coturnix-12.2%) applied across all the Phasianidae and Meleagrididae in the sample and also assumed that Ostrich, Rhea and Roadrunner for which there was no information had values equivalent to Tinamou (13%). The values for reptiles were derived after Ruben (1991). Note-the data used to generate the estimates for reptiles are exactly the same as those used by Ruben to calculate the power capacities of reptile muscles and the estimates I generated are exactly the same as his estimates. Also note that by taking performance of reptiles at C the estimate of their muscle power capacity is maximized. Even excluding the two low estimates in the reptile data set the muscle power generating capacities are not significantly different (two sample r-test, t = 1.85, P = 0.09). Ncr BM Max speed Power MS power Species (rnls, 02/kg 11m I) (kg) (rn/s I) (W.kg,) (W.kg-' muscle) Birds Ostrich Rhea Tinamou P. Quail Bobwhite Chuckar G. fowl Turkey Plover Roadrunner Mean (birds) SD (N = 10) Reptiles Amphibolurus Cnemidophorus Dipsosaurus Sceloporus Uma Trachydosaurus Ctenosaura Tiliqua Egernia Mean (reptiles) SD (N = 7) Mean (excluding Trachydosaurus and Tiliqua for which the estimate of muscle size may be erroneous: John-Alder et al., 1986) SD curve (see Fig. la). The difference between these two lines is the metabolic power which is lost as heat during the process of muscle contraction, and the ratio of the two is the efficiency. Over these low speeds muscle efficiency is low and constant (Fig. Ib). At some relatively low speed (marked as A in Fig. la) the reptile reaches its aerobic capacity (see Bennett, 1991). At greater speeds the required metabolic power input continues to increase but the aerobic power input remains constant at the maximum capacity. The difference is met by anerobic power input. Since mechanical power output continues to increase as a linear function of

4 NOTES AND COMMENTS 339 velocity but aerobic input remains constant there is an apparent exponential increase in the efficiency of muscle contraction when one calculates efficiency as the ratio of aerobic input and mechanical power output (Fig. Ib). At very high speeds, where the mechanical power output and aerobic input converge, apparent efficiency will in fact approach or even exceed 100% (point B on Fig. la). However, it is clear that this increase in efficiency is not real since the total costs of contraction (aerobic plus anerobic) continue to diverge from the mechanical requirements and actual muscle efficiency remains low (Fig. Ib dotted line), even at the animal's maximum burst speed. It is clear from Figure Ia that when one extrapolates the net cost oftransport line to the burst speed the prediction includes both the aerobic and anerobic components and thus also reflects the action of muscle working at low efficiency. The appropriate comparison to evaluate the flight performance of Archaeopteryx therefore is between available metabolic power derived using the extrapolation method and the empirically predicted metabolic power input requirements offlying animals (as performed here) and not the mechanical power output requirements which infer 100% muscle efficiency(as in Ruben, 1991). Second, I shall consider the comparison ofthe power generating capacities of avian compared to reptilian muscle. The methods employed to calculate these capacities by Ruben (1991, Fig. 4) are different for the two groups and not necessarily equivalent. The measurement of muscle power for mammals and birds reflects in vitro estimates, while the estimate for reptiles is based on the balance between observed locomotory metabolic energy expenditures divided by the muscle mass used in generating these expenditures. In the former case mechanical power generatingcapacitiesof the muscle are measured while in the latter oxygen consumption associated with generation ofsuch power is measured. These measures are not equivalent. Oxygen consumption is a measure of the total power input of the animal. Not all ofthis power appears as mechanical power output because muscles are not completely efficient and a substantial proportion of the metabolic power input appears as heat. The measurements on birds reflect mechanical power output and those on reptiles metabolic power input. If the same method for deriving the power input of reptilian muscle used by Ruben (1991) is used to measure the power input of avian muscle the power estimates are similar (Table I). These latter estimates for birds (c. 350 W.kg ') arc also close to the suggested aerobic capacities of avian flight muscle based on the volume density of mitochondria (Turner and Butler. 1988), which are around 300 W.kg '. Ifwe combine the estimates of metabolic power generating capacity of avian muscle with the known mass of the flight muscles ofa modem bird we can estimate the power available to support flight and compare this with the empirical prediction of metabolic flight energy requirements. For a pigeon (Columba lil'ia) this calculation suggests the flight muscles weighing 22.1% (Marden, 1987) of the 0.4 kg body mass. and using 354 W.kg, (Table I) could generate 31.3 Watts (SE = 5.9) of metabolic power. This compares very closely with an empirically predicted metabolic power requirement for a 0.4 kg flying animal of 29. I Watts (Speakman and Racey, 1991). Therefore the same approach adopted here which indicates that Archaeopteryx could not fly in sustained flapping flight also confirms that pigeons can. Furthermore, since my analysis indicates much closer metabolic capacities of avian and reptilian muscle than suggested by Ruben (1991). Archaeopteryx, with a flight muscle comprising only 9% of body mass, could not take off from the ground from a stationary position. ACKNOWLEDGMENTS I am most grateful to J. P. Hayes, G. C. Hays, A. Mackay. P. A. Racey, P. I. Webb. two anonymous referees, and especially J. Ruben for their constructive comments on earlier versions of this manuscript. LITERATURE CITED BENNETT, A. F The thermal dependence of lizard behaviour. Anim. Behav. 28: The evolution ofactivity capacity. J. Exp. BioI. 160:1-23. DODSON, P Conference report. International ArchaeopteryxConference. J. Vert. Palaeon. 5: I FEDAK, M. A., AND H. J. SEEHERMAN Reappraisal of energetics oflocomotion shows identical costs in bipeds and quadrupeds including ostrich and horse. Nature 282: GARLAND. T The relation between maximal running speed and body mass in terrestrial mammals. J. Zool. Lond. 199: Physiological correlates oflocomotory performance in a lizard: An allometric approach. Am. J. Physiol. 247:R806-8 I Ontogenetic and individual variation in size, shape and speed in the Australian agamid lizard Amphibolurus nuchalis. J. Zool. Lond. 207: HARTMAN. F. A Locomotor mechanisms of birds. Smithsonian Misc. Collections 143: JOHN-ALDER, H. B., T. GARLAND, AND A. F. BENNETT Locomotory capacities, oxygen consumption and the cost oflocomotion ofthe shingle-back lizard (Trachydosaurus rugosus). Physiol. Zool. 53: MARDEN. J. H Maximum lift production during take off in flying animals. J. Exp. BioI. 130: OSTROM, J. H Archaeopteryx and the origin of flight. Q. Rev. Biol. 49: Archaeopteryx and the origin of birds. BioI. J. Linn. Soc. 8: The meaning of Archaeopteryx, pp In M. K. Hecht et al. (eds.), The Beginnings of Birds. Freunde des Jura-Museums Eichstatt. PENNYCUICK. C. J Bird Flight Performance. Oxford University Press, Oxford, UK. RAYNER, J. M. V Cursorial gliding in protobirds, pp In M. K. Hecht, J. H. Ostrom, G. Viohl. and P. Wellnhofer(eds.), The Beginnings of Birds. Freunde des Jura Museums Eichstatt, RUBEN, J Reptilian physiology and the flight capacity ofarchaeopteryx. Evolution 45: SMIT. H.. J. M. AMELINK-KoUTSTAAL, J. VUVERBERG,

5 340 NOTES AND COMMENTS AND J. C. VON VAUPEL-KLEIN Oxygen consumption and efficiency of swimming goldfish. Compo Biochem. Physiol. 39A: SPEAKMAN, J. R.. AND P. A. RACEY No cost of echolocation for bats in flight. Nature 350: STEVENSON, R. D., AND R. K. JOSEPHSON Effects of operating frequency and temperature on mechanical power output from moth flight muscle. J. Exp. BioI. 149: TURNER. D. L., AND P. J. BUTLER The aerobic capacity of locomotor muscles in the tufted duck (Aythya fuligula). J. Exp. BioI. 135: WELLNHOFER, P Archaeopteryx. Sci. Am Y ALDEN, D. W The flying ability of Archaeopteryx. Ibis 114: Corresponding Editor: C. Hickman ANNOUNCEMENTS NATO ADVANCED STUDIES INSTITUTE, ADVANCES IN MORPHOMETRICS 11 Ciocco, Tuscany, Italy; July 18-30, 1993 The Institute will be concerned with the size and shape of organisms using geometric morphometries. It will cover theory and practice, with tutorials on coordinate data, multivariate statistics for morphometries, and outline methods. Applications to systematics and evolution will be emphasized. Time will be divided between discussions and hands-on computer work. Sixty participants will be selected among applicants from NATO and eastern Europe. Limited financial aid is available. Applications close April I, II Ciocco is a resort north of Lucca and Pisa. Cost for double room per person is 99,500 Italian Lira per day (about $800 for 12 days) including full board and travel to and from Pisa International Airport. Correspondence: Leslie F. Marcus, Department of Invertebrates, American Museum of Natural History, Central Park West at 79th Street, New York, NY Tel: , FAX: -5495, LAMQC@CUNYVM.BITNET or LAMQC@CUNYVM.CUNY.EDU. FIFfH ANNUAL MEETING OF THE HUMAN BEHAVIOR AND EVOLUTION SOCIETY Binghamton, New York; August 4-8, 1993 The fifth annual meeting of the Human Behavior and Evolution Society will be held on the campus ofbinghamton University. The society promotes scientific discourse in all disciplines by researchers who use the theory and methods of evolutionary biology to study humans. Research on nonhuman species is also welcome when it addresses general issues that are important to human evolution. Invited speakers include George C. Williams (keynote), J. Michael Bailey, Leda Cosmides and John Tooby, Martin Daly and Margo Wilson, William Durham, Harry Harpending, and Elliott Sober. Organized symposia include "Evolutionary approaches to cognition," "Evolutionary approaches to morality," and "Evolution and culture." Deadline for submission of abstracts is May I, Correspondence: David Sloan Wilson, Human Behavior and Evolutionary Society, Department of Biological Sciences, Binghamton University, Binghamton, NY ; Phone: (607) , FAX: (607) , DWILSON@BINGVAXA.BITNET.