5D Library › Scaling Laws, Allometric Growth and the Statistical Physics of Organisational Metabolism · 中文

Chapter 1 Kleiber’s Law and Fractal Networks: The Three-Quarter Power and Its Disputes

A straight line with a slope near 3/4

If metabolic rate were proportional to body mass, each kilogram of an elephant would burn as much energy as each kilogram of a mouse; if it were proportional to body surface area (volume grows as the cube of length, area as the square), metabolic rate would be proportional to mass to the 2/3 power. The nineteenth-century “surface law” took the second view. In 1932 Kleiber compared 13 groups of animals, from pigeons and rats to steers, and found that the 3/4 power of body mass was a better unit for comparison than surface area (Kleiber 1932); his 1947 review discussed and popularised this unit further (Kleiber 1947). Since then B∝M3/4B \propto M^{3/4} has been called Kleiber’s law.

Taking logarithms of B=B0MbB = B_0 M^{b} gives log⁡B=log⁡B0+blog⁡M\log B = \log B_0 + b \log M: on logarithmic axes a power law is a straight line whose slope is the exponent bb. The dispute over “3/4 or 2/3” is a dispute about the slope of this line. The two values are close but mean different things: with b=3/4b = 3/4, a ten-thousandfold increase in mass reduces the metabolic rate per unit mass to one tenth; with b=2/3b = 2/3, to about one twenty-second.

The slope depends on which part you look at

Evidence

Large-sample results. White and Seymour (2003) compiled basal metabolic rates for 619 mammal species from 19 orders, spanning five orders of magnitude in mass, and concluded that the data do not support 3/4: basal metabolic rate is proportional to mass to the 2/3 power (White and Seymour 2003). Dodds, Rothman and Weitz (2001) took 2/3 as a testable null hypothesis: in Heusner’s compilation the exponent for the 357 mammal species of 10 kg or less was 0.668 (95 % confidence interval 0.643 to 0.693), for all 391 species 0.710 (0.689 to 0.731), rising systematically as larger animals were included; in Kleiber’s own 13 groups it was 0.667 below 10 kg and 0.754 above. They judged that the theories then offered for 3/4 all rested on unconvincing assumptions (Dodds et al. 2001). Glazier’s 2005 review showed that metabolic exponents range roughly from 2/3 to 1 across taxa, activity levels and physiological states, so that a “universal 3/4” is more of an average impression (Glazier 2005). Kolokotrones and colleagues (2010) pointed out that the mammalian data are themselves curved (convex) on logarithmic axes, so that a straight-line fit naturally gives a slope that depends on the range of masses (Kolokotrones et al. 2010).

If the true relation is curved, the question “what is the exponent?” must be qualified by “over which range of mass?”. The program below builds a synthetic data set: 400 “species” from 10 g to 4 tonnes, whose true curve is slightly convex on logarithmic axes, with a local slope rising from about 0.60 to 0.82. The shape is only illustrative; its direction matches that reported by Dodds and colleagues and by Kolokotrones and colleagues, but it is not fitted to any real data.

Calculation

For each range of mass, an ordinary least-squares (OLS) regression gives the slope and its 95 % confidence interval, and the program checks whether 2/3 and 3/4 lie inside it; it also gives the reduced major axis (RMA) slope, the OLS slope divided by the correlation coefficient, another fitting method that Chapter 6 will use again. The last two lines do the “quarter-power” bookkeeping for heart rate and lifespan.

# Kleiber's law on SYNTHETIC data: 400 "species" from 10 g to 4,000 kg. The true curve
# is slightly convex in log-log (local exponent rises from about 0.60 to 0.82), an
# ILLUSTRATIVE shape in the direction reported by Dodds et al. 2001 and Kolokotrones et
# al. 2010; it is not fitted to any real data set. Question: which exponent does a
# straight-line fit give, and does it depend on the mass range and the method?
import math, random
from statistics import NormalDist, mean
rng = random.Random(1932)
C0, C1, C2 = 0.53, 0.68, 0.02           # log10 B = C0 + C1 x + C2 x^2, x = log10(M/kg)

def true_log_b(x):
    return C0 + C1 * x + C2 * x * x

def fit(xs, ys):
    """OLS slope, its standard error, correlation r and the RMA (major-axis) slope."""
    mx, my = mean(xs), mean(ys)
    sxx = sum((x - mx) ** 2 for x in xs); syy = sum((y - my) ** 2 for y in ys)
    sxy = sum((x - mx) * (y - my) for x, y in zip(xs, ys))
    b = sxy / sxx; r = sxy / math.sqrt(sxx * syy)
    res = sum((y - my - b * (x - mx)) ** 2 for x, y in zip(xs, ys))
    se = math.sqrt(res / (len(xs) - 2) / sxx)
    return b, se, r, b / abs(r)

if __name__ == "__main__":
    xs = [rng.uniform(-2.0, 3.6) for _ in range(400)]
    ys = [true_log_b(x) + rng.gauss(0, 0.12) for x in xs]
    print("local exponent of the true curve: 10 g %.2f | 1 kg %.2f | 1 t %.2f" %
          tuple(C1 + 2 * C2 * x for x in (-2, 0, 3)))
    z = NormalDist().inv_cdf(0.975)
    print("mass range      n  OLS slope  95 % CI       RMA slope  2/3 in CI  3/4 in CI")
    for lab, lo, hi in (("< 10 kg", -2, 1), ("all", -2, 3.6), ("> 10 kg", 1, 3.6),
                        ("0.1 - 100 kg", -1, 2)):
        sub = [(x, y) for x, y in zip(xs, ys) if lo <= x < hi]
        b, se, r, rma = fit([p[0] for p in sub], [p[1] for p in sub])
        ins = lambda v: "yes" if abs(v - b) <= z * se else "no"
        print(f"{lab:13s} {len(sub):4d}   {b:.3f}   {b - z * se:.3f}-{b + z * se:.3f}"
              f"    {rma:.3f}      {ins(2 / 3):3s}        {ins(0.75)}")
    # Quarter powers: heart rate ~ M^-1/4 and lifespan ~ M^+1/4, so their product (beats
    # per lifetime) does not depend on mass. Levine 1997: mammals average about 7.3e8
    # beats (SD 5.6e8). A human at 70 beats per minute for 80 years:
    human = 70 * 60 * 24 * 365.25 * 80
    print(f"exponent of beats per lifetime: -1/4 + 1/4 = {-0.25 + 0.25:.2f}")
    print(f"human, 70 /min for 80 years: {human:.2e} beats = {human / 7.3e8:.1f} x the"
          " mammalian average")
# local exponent of the true curve: 10 g 0.60 | 1 kg 0.68 | 1 t 0.80
# mass range      n  OLS slope  95 % CI       RMA slope  2/3 in CI  3/4 in CI
# < 10 kg        224   0.652   0.635-0.670    0.665      yes        no
# all            400   0.707   0.699-0.715    0.711      no         no
# > 10 kg        176   0.770   0.746-0.795    0.788      no         yes
# 0.1 - 100 kg   212   0.695   0.674-0.715    0.711      no         no
# exponent of beats per lifetime: -1/4 + 1/4 = 0.00
# human, 70 /min for 80 years: 2.95e+09 beats = 4.0 x the mammalian average

From the same data, species under 10 kg give 0.652, with an interval that contains 2/3 but not 3/4; species over 10 kg give 0.770, with an interval that contains 3/4 but not 2/3; all species give 0.707, and neither value lies in the interval. The RMA slopes are slightly larger but change no conclusion. So when two studies report 2/3 and 3/4, they need not contradict each other; they may simply have chosen different species. This does not make the dispute empty: White and Seymour argued that differences in body temperature between species, and large herbivores measured when not in a fasting, basal state, both push the fitted exponent upwards (White and Seymour 2003); what counts as a “basal” measurement is itself part of the dispute.

Left: synthetic data (grey points) and straight-line fits for three ranges of mass; the slope varies with the range. Right: local exponent of the West–Brown–Enquist network against the number of branching levels. With area-preserving branching only, the exponent approaches 3/4 from below (blue line); with the last 15 levels in Poiseuille flow, from above (gold dashed line); with Poiseuille flow throughout, it tends to 1 (purple dotted line).

The fractal network of West, Brown and Enquist

Evidence

Three assumptions of the WBE model. West, Brown and Enquist proposed in Science in 1997 that organisms distribute resources through hierarchically branching networks (blood vessels, tracheae, plant vascular tissue), and assumed: (1) the network is space-filling, each level of branching serving the whole volume it supplies; (2) the terminal units (capillaries, for example) are the same size whatever the size of the organism; (3) natural selection has minimised the energy spent on transport. From these assumptions they derived that metabolic rate is proportional to mass to the 3/4 power, and the same model gave other quarter-power relations, for example an aortic radius proportional to roughly M3/8M^{3/8} and a heart rate proportional to M−1/4M^{-1/4} (West et al. 1997).

Calculation

Let each vessel divide into nn daughters over NN levels, giving nNn^N terminal capillaries. Because the terminal units are invariant, metabolic rate B∝nNB \propto n^N. Space filling requires the length ratio between successive levels to be lk+1/lk=n−1/3l_{k+1}/l_k = n^{-1/3}. In large vessels dominated by pulsatile flow, impedance matching requires cross-sectional area to be preserved, a radius ratio rk+1/rk=n−1/2r_{k+1}/r_k = n^{-1/2}; in small vessels dominated by viscosity (Poiseuille flow), energy minimisation gives rk+1/rk=n−1/3r_{k+1}/r_k = n^{-1/3}. Blood volume is V=∑knkπrk2lkV = \sum_k n^k \pi r_k^2 l_k, and body mass is assumed to satisfy M∝VM \propto V.

If all branching is area-preserving, the level jj steps above the capillaries contains nN−jn^{N-j} vessels, each of volume ∝nj⋅nj/3\propto n^{j} \cdot n^{j/3}, so the total volume of the level is ∝nNnj/3\propto n^N n^{j/3}, and summing gives V∝nN⋅nN/3V \propto n^N \cdot n^{N/3} for large NN. Hence V∝n4N/3V \propto n^{4N/3} and B∝nN∝V3/4B \propto n^N \propto V^{3/4}. This is where 3/4 comes from: the extra factor nN/3n^{N/3} in the ratio of volume to the number of terminal units comes from the length scaling required by space filling.

The program below computes networks of finite size level by level and gives the local exponent dln⁡B/dln⁡M\mathrm{d}\ln B/\mathrm{d}\ln M between neighbouring sizes.

# The West-Brown-Enquist (1997) branching network, computed level by level. Each vessel
# splits into n = 2 daughters. Capillaries (the last level) are the same in every animal,
# so metabolic rate B ~ number of capillaries = n^N. Space filling: length ratio
# l(k+1)/l(k) = n^(-1/3). Radius ratio: area-preserving n^(-1/2) in large pulsatile
# vessels, area-increasing n^(-1/3) in small Poiseuille vessels. Body mass M ~ blood
# volume V = sum over levels of n^k * r_k^2 * l_k. Units: capillary r = l = 1.
import math
n = 2

def volume(levels, poiseuille=0):
    """Blood volume of a tree with 'levels' branchings; the last 'poiseuille' ones are
    area-increasing (radius ratio n^-1/3), the rest area-preserving (n^-1/2)."""
    v, r, l = 0.0, 1.0, 1.0
    for k in range(levels, -1, -1):           # walk from the capillaries up to the aorta
        v += n ** k * r * r * l
        up = levels - k                       # number of steps above the capillaries
        r *= n ** (1 / 3) if up < poiseuille else n ** 0.5
        l *= n ** (1 / 3)
    return v

def local_exponent(levels, poiseuille=0):
    """d ln B / d ln M between trees with levels and levels + 1 branchings."""
    b0, b1 = n ** levels, n ** (levels + 1)
    return math.log(b1 / b0) / math.log(volume(levels + 1, poiseuille)
                                         / volume(levels, poiseuille))

if __name__ == "__main__":
    print("levels N  capillaries  local exponent d ln B / d ln M")
    print("          (= B)        area-preserving  last 15 Poiseuille  all Poiseuille")
    for lv in (6, 12, 18, 24, 30, 40, 60, 200):
        print(f"  {lv:4d}   {float(n ** lv):9.2e}       {local_exponent(lv):.3f}"
              f"            {local_exponent(lv, 15):.3f}              "
              f"{local_exponent(lv, lv + 1):.3f}")
    # Average exponent over an ILLUSTRATIVE span of N = 16 to 30 branchings (B varies by
    # 2^14, about four orders of magnitude), for different numbers of Poiseuille levels
    lo, hi = 16, 30
    for p in (0, 6, 10, 15):
        e = (hi - lo) * math.log(n) / math.log(volume(hi, p) / volume(lo, p))
        print(f"average exponent, N = {lo} to {hi}, last {p:2d} levels Poiseuille:"
              f" {e:.3f}")
# levels N  capillaries  local exponent d ln B / d ln M
#           (= B)        area-preserving  last 15 Poiseuille  all Poiseuille
#      6    6.40e+01       0.712            0.838              0.838
#     12    4.10e+03       0.741            0.903              0.903
#     18    2.62e+05       0.748            0.859              0.931
#     24    1.68e+07       0.749            0.790              0.946
#     30    1.07e+09       0.750            0.761              0.956
#     40    1.10e+12       0.750            0.751              0.966
#     60    1.15e+18       0.750            0.750              0.977
#    200    1.61e+60       0.750            0.750              0.993
# average exponent, N = 16 to 30, last  0 levels Poiseuille: 0.749
# average exponent, N = 16 to 30, last  6 levels Poiseuille: 0.752
# average exponent, N = 16 to 30, last 10 levels Poiseuille: 0.766
# average exponent, N = 16 to 30, last 15 levels Poiseuille: 0.811

The results show the conditions under which 3/4 holds. With area-preserving branching only, a small network of 6 levels has an exponent of only 0.712, 12 levels 0.741, and the value settles at 0.750 only beyond 30 levels: finite networks approach 3/4 from below. If all branching is of the Poiseuille type, every level has the same total volume, V∝(N+1)nNV \propto (N+1) n^N, and the exponent tends to 1, with no quarter power; in the WBE model, therefore, 3/4 depends on pulsatile flow in the large vessels. Real circulatory systems contain both regimes: with the last 15 levels of Poiseuille type, the local exponent is 0.903 at 12 levels and 0.859 at 18, 0.790 at 24, and approaches 0.750 only beyond 40 levels. Over the illustrative range of 16 to 30 levels, the average exponent is 0.811 (0.752 if only the last 6 levels are of Poiseuille type). The level counts and regimes here are illustrative; Savage, Deeds and Fontana (2008) derived the finite-size corrections with more detailed parameters and concluded that over the actual range of mammalian sizes the WBE model predicts an exponent of about 0.81, at odds with the data (Savage et al. 2008).

Note

Criticisms of the model. Kozłowski and Konarzewski (2004) questioned whether the WBE derivation is mathematically consistent and whether its assumptions are biologically realistic, for instance whether terminal units are really invariant and whether body mass is proportional to blood volume (Kozłowski and Konarzewski 2004). Dodds and colleagues argued that none of the theories, network models included, convincingly shows why the exponent should be 3/4 rather than 2/3 (Dodds et al. 2001). This book’s position is that the WBE model is a clear, computable hypothesis: it shows that a fractal network with invariant terminal units produces sublinear scaling, but whether it is the true cause of the metabolic exponent is still debated.

A lifetime of heartbeats

Evidence

An approximate invariant: heartbeats per lifetime. If heart rate is proportional to M−1/4M^{-1/4} and lifespan to M1/4M^{1/4}, their product, the number of heartbeats in a lifetime, does not depend on mass. Levine (1997) compiled resting heart rates and life expectancies of mammals and found an average of about 7.3×1087.3 \times 10^8 beats per lifetime, with a standard deviation of about 5.6×1085.6 \times 10^8, the values lying within an order of magnitude of one another; the mean the abstract gives when the data are taken together is about 10×10810 \times 10^8; he also noted that whether slowing the human heart rate would extend life remains an open question (Levine 1997). Austad and Fischer (1991) found that in bats and marsupials the relation between lifespan and metabolism departs markedly from that of other mammals: bats live much longer than non-flying mammals of the same mass and do not “use up” the same lifetime quota of energy (Austad and Fischer 1991).

The last two lines of the program do this bookkeeping: the exponent is −1/4+1/4=0-1/4 + 1/4 = 0. At 70 beats a minute for 80 years, a human has about 2.95×1092.95 \times 10^9 heartbeats in a lifetime, 4.0 times the mammalian average reported by Levine. The framework’s “about 1 to 1.5 billion” therefore needs qualifying: the mammalian average compiled by Levine is about 0.7 billion and his pooled estimate about 1 billion, acceptable as an order of magnitude, but species differ by up to a factor of ten, so this is no narrow “constant”; and humans, bats and birds are clear exceptions. This is a statistical regularity of comparative biology, not a “heartbeat quota” for each person, and it certainly does not imply that slowing the heart lengthens life.

Hypothesis · not established

The invariant front-line unit hypothesis (not yet established). The key assumption of the WBE model is that terminal units do not grow with the organism. The author conjectures that in firms that grow only by adding front-line teams of constant size (say 5 to 8 people each), the number of managerial and support staff grows with total headcount with an exponent below 1, whereas in firms whose front-line teams keep growing with size the exponent is near or above 1. The testable prediction is that, in pre-registered panel data on many firms, the difference between the two exponents is negative. If the 95 % confidence interval of the difference contains no negative values, the hypothesis is refuted.

Summary

Kleiber’s law is a line that holds approximately on logarithmic axes, but its slope depends on the species, the range of mass, the physiological state and the fitting method, and fitting a single straight line may hide curvature. The WBE fractal-network model shows that a space-filling branching network with invariant terminal units produces sublinear scaling; 3/4 holds exactly only for an infinitely large area-preserving network, and finite size and the Poiseuille regime move the exponent away from it. The “invariant” number of heartbeats per lifetime is only a rough statistical trend: Levine’s compiled average is about 0.7 billion and the order of magnitude about 1 billion, with species differing by up to tenfold, not a fixed quota of 1 to 1.5 billion. The next chapter asks: if supply grows sublinearly and maintenance linearly, where does growth stop?

References

Austad, Steven N., and Kathleen E. Fischer. 1991. “Mammalian Aging, Metabolism, and Ecology: Evidence from the Bats and Marsupials.” Journal of Gerontology 46 (2): B47–53. https://doi.org/10.1093/geronj/46.2.B47.
Dodds, Peter Sheridan, Daniel H. Rothman, and Joshua S. Weitz. 2001. “Re-Examination of the ‘3/4-Law’ of Metabolism.” Journal of Theoretical Biology 209 (1): 9–27. https://doi.org/10.1006/jtbi.2000.2238.
Glazier, Douglas S. 2005. “Beyond the ‘3/4-Power Law’: Variation in the Intra- and Interspecific Scaling of Metabolic Rate in Animals.” Biological Reviews 80 (4): 611–62. https://doi.org/10.1017/S1464793105006834.
Kleiber, Max. 1932. “Body Size and Metabolism.” Hilgardia 6 (11): 315–53. https://doi.org/10.3733/hilg.v06n11p315.
Kleiber, Max. 1947. “Body Size and Metabolic Rate.” Physiological Reviews 27 (4): 511–41. https://doi.org/10.1152/physrev.1947.27.4.511.
Kolokotrones, Tom, Van Savage, Eric J. Deeds, and Walter Fontana. 2010. “Curvature in Metabolic Scaling.” Nature 464 (7289): 753–56. https://doi.org/10.1038/nature08920.
Kozłowski, Jan, and Marek Konarzewski. 2004. “Is West, Brown and Enquist’s Model of Allometric Scaling Mathematically Correct and Biologically Relevant?” Functional Ecology 18 (2): 283–89. https://doi.org/10.1111/j.0269-8463.2004.00830.x.
Levine, Herbert J. 1997. “Rest Heart Rate and Life Expectancy.” Journal of the American College of Cardiology 30 (4): 1104–6. https://doi.org/10.1016/S0735-1097(97)00246-5.
Savage, Van M., Eric J. Deeds, and Walter Fontana. 2008. “Sizing up Allometric Scaling Theory.” PLoS Computational Biology 4 (9): e1000171. https://doi.org/10.1371/journal.pcbi.1000171.
West, Geoffrey B., James H. Brown, and Brian J. Enquist. 1997. “A General Model for the Origin of Allometric Scaling Laws in Biology.” Science 276 (5309): 122–26. https://doi.org/10.1126/science.276.5309.122.
White, Craig R., and Roger S. Seymour. 2003. “Mammalian Basal Metabolic Rate Is Proportional to Body Mass2/3.” Proceedings of the National Academy of Sciences of the USA 100 (7): 4046–49. https://doi.org/10.1073/pnas.0436428100.

Fan, Y. (2026). Scaling Laws, Allometric Growth and the Statistical Physics of Organisational Metabolism: From Kleiber's Three-Quarter Power Law and Fractal Supply Networks to the Mortality of Firms, Read Beside Classical Chinese Organisational Thought [规模法则、异速生长与组织新陈代谢的统计物理] (Version 1.0.0). 5D Library. https://doi.org/10.5281/zenodo.23788553 · Licensed under Creative Commons Attribution-NonCommercial-NoDerivatives 4.0 International (CC BY-NC-ND 4.0).