brief communications Pattern and intensity of physical activity Keeping moderately active is the best way to boost total daily energy expenditure. low level of physical activity is a per- 2.2 taining energy balance and may lose weight; vasive feature of our modern lifestyle. PAL is about 1.5 or 2.1 for sedentary or very Attempts to counteract the negative active people, respectively5. My results indi- A 2.0 effects of sedentary living include introduc- cating that short periods of vigorous activi- ing bouts of high-intensity exercise, but I ty do not have much impact on PAL in the 1.8 show here that it can be more effective to PAL normal population are borne out by studies increase the amount of time spent on activ- on obese patients: adding exercise to an ities of moderate intensity while reducing 1.6 energy-restricted diet does not further periods of inactivity during waking hours. increase weight loss6,7 because the costs of 1.4 Physical inactivity may be associated not 02040 60 80 100 the extra activity are compensated by a only with being overweight but also with Fraction of daytime (%) reduction of energy spent on physical activ- disorders such as coronary heart disease or ity outside the training sessions8. type-2 diabetes1, which may be prevented Figure 1 Physical activity levels (PAL) as a function of the fraction My results show, however, that the pro- by exercising regularly2. Consequently, of daytime hours spent by 30 healthy subjects (with body-mass portion of time distributed between activi- sedentary people often include spells of vig- indices within the normal range) on activities of low (filled circles), ties of low and moderate intensity is what orous exercise in an otherwise inactive and moderate (circles) and high (squares) exercise intensity. Linear influences the total energy expenditure and unhealthy lifestyle. I have tested whether regressions are shown for low- and moderate-intensity activities: so determines the value of PAL. Subjects this is the most effective way to enhance subjects spending more time on moderate-intensity exercise and wanting to increase their metabolic rate people’s total daily activity levels by deter- less on low-intensity activity can improve their PAL values. Time should exchange low-intensity activities mining how these are influenced by exercise spent on high-intensity exercise does not appear to influence total such as sitting in front of a screen for mod- duration and intensity. energy expenditure. erate-intensity activities such as walking or Healthy, non-obese adults (14 women cycling. Moderate-intensity activities are and 16 men; 2755 yr; 24.152.3 kg m–2) Calculated PAL values fell in the range better tolerated than high-intensity activi- participated in the study3. Total energy 1.51–2.04 for all subjects, apart from one ties, especially by the middle-aged or obese. expenditure was measured by monitoring woman whose PAL value was 2.57; her data An improvement in overall activity levels their metabolism with doubly labelled were excluded as an outlier. Three subjects should bring important health benefits as water over a two-week period, the optimal reported not wearing the Tracmor while subjects expend more energy and reduce observation interval for the biological performing intensive sports activities, so their risk of running into positive energy half-lives of the isotopes. The physical the time spent on high-intensity activity by balance and gaining weight. activity level (PAL) is defined as the factor these subjects was corrected accordingly. Klaas R. Westerterp by which total energy expenditure exceeds In a multiple regression analysis with the Department of Human Biology, Maastricht resting energy expenditure, measured fraction of time spent on activities of mod- University, 6200 MD Maastricht, The Netherlands during an overnight stay in a respiration erate and high intensity, only moderate- e-mail: [email protected] chamber. intensity activity came out as a significant 1. Booth, F. W., Gordon, S. E., Carlson, C. J. & Hamilton, M. T. To determine the amount of body move- predictor of PAL (r 240.51, P*0.0001). J. Appl. Physiol. 88, 774–787 (2000). ment over 1-min intervals, I recorded Subjects spending relatively more time on 2. Grundy, S. M. et al. Med. Sci. Sports Exerc. 31 (suppl.), 502–508 (1999). movements over the first of the two weeks moderate-intensity activity and therefore 3. Westerterp, K. R. et al. Br. J. Nutr. 76, 785–795 (1996). with a ‘Tracmor’ portable motion sensor. less on low-intensity activity had a higher 4. Bouten, C. V. C., Verboeket-van de Venne, W. P. H. G., This consists of a body-fixed triaxial PAL value (Fig. 1). There was, however, no Westerterp, K. R., Verduin, M. & Janssen, J. D. J. Appl. Physiol. accelerometer and a data unit for on-line relation between PAL value and the time 81, 1019–1026 (1996). 5. Westerterp, K. R.. Am. J. Clin. Nutr. 68 (suppl.), 970–974 registration, processing and storage of spent on just high-intensity activity, pre- (1998). acceleration signals. The triaxial accelerom- sumably because this was limited by its 6. Ballor, D. L. & Poehlman, E. T. Eur. J. Appl. Physiol. 71, 535–542 eter (5023028 mm; 16 g) consists of three nature to being relatively short. (1995). uniaxial piezoresistive accelerometers and In the general population, PAL ranges 7. Garrow, J. S. & Summerbell, C. D. Eur. J. Clin. Nutr. 49, 1–10 (1995). registers accelerations in three orthogonal between 1.2 and 2.2–2.5. At PAL values of 8. Kempen, K. P. G., Saris, W. H. M. & Westerterp, K. R. Am. J. directions as the body moves. Acceleration about 2.5, subjects have problems main- Clin. Nutr. 62, 722–729 (1995). signals are processed to obtain the sum of the rectified and integrated acceleration curves from all three measurement direc- Forensic palaeontology between birds and extinct non-avian tions. The integration time is set at 1 min dinosaurs2,3. But Archaeoraptor was revealed and the final output is expressed as counts The Archaeoraptor to be a forgery in which bones of a primitive per minute (c.p.m.)4. bird and a non-flying dromaeosaurid Using data from Tracmor recordings in forgery dinosaur had been combined4–6. Here we combination with activity details recorded he Archaeoraptor fossil was announced use high-resolution X-ray computed in diaries, three activity categories were as a ‘missing link’ and purported to be tomography (CT)7 to determine the nature defined: low intensity (lying, sitting and Tpossibly the best evidence since and extent of the forgery, as well as how it standing; Tracmor output under 1,000 Archaeopteryx that birds did, in fact, evolve was built, by imaging the fracture pattern c.p.m.), moderate intensity (walking and from certain types of carnivorous and distribution of materials through the cycling; output 1,000–3,000 c.p.m.); and dinosaur1. It reportedly came from Early entire specimen. high intensity (housework, gymnastics and Cretaceous beds of China that have pro- The Archaeoraptor specimen, which was sport; output greater than 3,000 c.p.m.). duced other spectacular fossils transitional reportedly collected from the Early Creta- NATURE | VOL 410 | 29 MARCH 2001 | www.nature.com © 2001 Macmillan Magazines Ltd 539 brief communications ceous Jiufotang Formation of Liaoning, was ab smuggled out of China and later sold in the HH 1b 1e GG United States on the commercial market4. 1d 5 1a FF Before repatriation , the specimen, which 1h A 1g was alleged to contain a single complete 1f 1n B 1c skeleton, was brought to us for CT analysis. 1m 1o EE The slab had obviously been reassem- C 1i 1k 1l DD bled from numerous broken pieces, and the 1j E skeleton presented a unique combination of D 1w 4a 1r 1t 1u 1v 4b 1p 1s features, including teeth, feathers, a more 4f 3 CC 4c 4g BB 2 flight-capable wing than Archaeopteryx, and 1q 8b AA 4d 6 4i Z F 7a 7b the tail of a non-flying dromaeosaur. The 8a 4h Y 9a O 4j X entire bone-bearing portion of the slab was 4e 9b H W 9c V G I scanned in 422 consecutive 1-mm-thick 9d 10 U 5a T slices, spaced at 0.9 mm (for the full dataset J 11 and description of scanning protocols, see S 5c 12b supplementary information). 12a 5e 5b Computed tomography alone cannot K 5d R distinguish genuine and fraudulent mis- 12c 13a Q takes, but it provides criteria for evaluating 13b whether pairs of adjacent pieces lie in nat- L P N ural relation to one another. These include M comparison of adjacent fracture-face geometries, thickness and density of adja- Figure 1 Two computer-generated models of the face of the Archaeoraptor slab as it was presented for computed tomography (CT) scan- cent pieces, cross-cutting relationships ning on 29 July 1999. The specimen was scanned from top to bottom in planes perpendicular to this view. a, Volumetric model generat- among four generations of fractures, the ed from superimposed CT slices. Light grey, bones; dark grey, slab materials; black, air. b, Map of the slab face, colour-coded to indicate distribution of grout, and continuity across the nature of the associations between its 88 constituent pieces. Red, associated bird bones; black, unverifiable ‘attached’ bones. 1a–w, fractures of bones, natural moulds of bones, Associated pieces lying in a natural position; 2,3, ‘left’ femur; 4a–j, ‘right’ and ‘left’ tibia/fibula (piece and counterpiece); 5a–e, ‘right’ and invertebrate burrows. The textural sig- foot/ankle (piece and counterpiece); 6, 7a,b, 8a–c, bone fragments; 9a–d, 10, 11, 12a–c, 13a,b, dromaeosaur tail pieces; A–HH, shims.