Scaling Laws: From Elephant Metabolism to City Growth

In the 1960s, the CIA's top-secret MKUltra project aimed to understand how drugs like LSD could alter human behavior. One hypothesis involved elephants, which are typically docile but can sometimes "snap." Researchers theorized this behavioral change might be triggered by a naturally occurring, LSD-like substance in their brains. If true, administering LSD to a docile elephant could replicate this behavior. The challenge was determining a dose large enough to cause a psychological reaction without causing harm.

The Tragic Case of Tusko the Elephant

LSD had never been given to an animal as large as an elephant. Researchers knew the safe dose for cats was around 0.3 milligrams. Assuming a linear scaling with mass, and knowing an elephant is roughly a thousand times the mass of a cat, they decided to administer a thousand times the dose.

In an experiment approved for Tusko, an Indian elephant at the Lincoln Park Zoo in Oklahoma, nearly 300 milligrams of LSD were injected. Within five minutes, Tusko trumpeted, collapsed, went into status epilepticus, and died shortly after despite attempts to revive him. The fatal error was the assumption that safe drug dosage scales linearly with mass, which it does not. This incident highlights that many phenomena do not scale as intuitively expected.

The Billion Heartbeat Mystery

Consider the Etruscan shrew, the smallest mammal, and the African bush elephant, the largest land mammal. Which has more heartbeats in its lifetime? Surprisingly, both have approximately one billion heartbeats. This pattern holds true for almost every mammal, regardless of size, environment, or lifespan. This observation, largely explored in Geoffrey West's book "Scale," suggests a fundamental biological constraint.

Knowing a mammal's mass allows for the prediction of numerous biological traits, from pulse rate and reproductive output to total lifespan. This scaling phenomenon also applies to cities, where population and location can predict average wages, patent filings, crime rates, disease prevalence, and even pedestrian walking speeds.

Metabolic Rate and the Surface Law

The key to understanding these scaling laws lies in metabolic rate—the energy an animal uses over time. Researchers initially assumed drug dosage was proportional to mass because they believed metabolic rate scaled linearly with mass. However, the speed at which an animal processes chemical compounds depends more on its metabolic rate than its mass.

While a cat requires about 250 kilocalories per day, an elephant, with a thousand times the mass and thus a thousand times the cells, would intuitively be expected to need 250,000 kilocalories. However, this leads to a problem: if an organism burns energy, it radiates heat through its surface.

To illustrate, imagine animals as perfect spheres. An elephant, with a thousand times the volume of a cat, has a radius only ten times larger (since volume scales with radius cubed). Its surface area, however, only increases by a factor of 100 (since surface area scales with radius squared). If an elephant generated a thousand times the heat but only had a hundred times the surface area to radiate it, it would overheat and die.

In 1838, French scientists proposed the "surface law," stating that metabolic rate (B) should scale proportionally to surface area (A). This implies that metabolic rate scales with mass to the two-thirds power (M^(2/3)). According to this law, an elephant a thousand times heavier than a cat should burn only a hundred times as many calories (25,000 instead of 250,000), and Tusko's appropriate LSD dose would have been 30 milligrams.

This concept is analogous to cooking a turkey: the cooking time doesn't scale linearly with weight but rather with the thickness of the bird, which relates to mass to the two-thirds power. Doubling the weight of a roast only requires about 60% longer cooking time, not 100%.

Kleiber's Law and the Quarter Power Scaling

These relationships are known as power laws, characterized by a straight line when plotted on a log-log graph, where the slope represents the exponent. A slope of one indicates linear scaling, less than one is sublinear, and greater than one is superlinear.

For nearly a century, the two-thirds exponent of the surface law was widely accepted for metabolic rate. However, in 1932, Swiss biologist Max Kleiber tested this by plotting the metabolic rates of various animals against their mass on a log-log plot. He found a straight line, but the slope was approximately three-quarters (M^(3/4)), not two-thirds. This became known as Kleiber's Law.

Kleiber's Law suggests that if an animal's mass doubles, its metabolic rate increases by about 68%, not 59% as predicted by the two-thirds law. According to Kleiber's Law, an elephant burns roughly 178 times as many calories as a cat (about 45,000 kilocalories), and Tusko's correct LSD dose would have been 53 milligrams—about a sixth of what he received.

Kleiber's original work was based on a small dataset of mammals. However, broader studies including mammals, birds, reptiles, and fish show the same three-quarters scaling relationship, though warm-blooded animals have higher base metabolic rates. Some even argue this relationship extends to single cells, spanning over 25 orders of magnitude. This implies an efficiency to being larger, as cells in larger creatures use proportionately less energy per unit of mass.

The WBE Theory: Explaining Quarter Power Laws

The mystery deepened as researchers found that brain size, growth rate, and blood pumped per minute also scale roughly as mass to the three-quarters. Other properties, like lifespan and blood circulation time, scale as mass to the one-quarter, while breathing and heart rate scale to the negative one-quarter. The consistent appearance of quarter-power exponents prompted the question: where do they come from?

In the 1990s, Brian Enquist, inspired by these scaling plots, teamed up with Professor James Brown and theoretical physicist Geoffrey West. They formed the West, Brown, and Enquist (WBE) Theory, seeking a compelling explanation for Kleiber's Law.

Their theory rests on three premises: 1. Space-filling networks: Resource distribution networks (like circulatory systems) must reach every cell. 2. Invariant terminal units: The thinnest segments of these networks (e.g., capillaries) have the same width regardless of organism size. An elephant simply has more of them. 3. Efficient design: Evolution has optimized these transport networks for efficiency.

An efficient network design involves branching, self-similar fractal structures. To minimize energy loss from blood reflections at branching points, the cross-sectional area of vessels must remain constant before and after branching. This leads to a fractal geometry, mirroring the actual structure of circulatory systems.

Mathematician Felix Hausdorff's work on fractal dimensions provides the link to quarter-power scaling. A space-filling fractal curve, like a crumpled piece of paper filling a 3D volume, has a Hausdorff dimension of 3.0. This means the surface area of the circulatory system effectively scales with its length cubed, allowing for a massive amount of membrane surface area to be packed in.

Since every cell needs to be served by the network, the volume around the network is proportional to the animal's mass. Given that volume is proportional to length to the fourth (due to the fractal nature of the network), length is proportional to mass to the one-quarter (M^(1/4)). Plugging this into the metabolic rate equation (which scales with length cubed), the metabolic rate is found to be proportional to mass to the three-quarters (M^(3/4)), precisely Kleiber's Law.

Predictions and Implications of WBE Theory

Published in 1997, WBE Theory makes specific predictions for 26 different biological scaling exponents, many of which are not multiples of a quarter. For example, the radius of an animal's aorta should scale with mass to the three-eighths (0.375), and lung area with mass to the 11/12ths (0.92). Observed data closely match these predictions (e.g., aorta radius 0.36, lung area 0.95).

The theory also explains other scaling laws. Heart rate, for instance, scales as metabolic rate divided by mass (B/M). Since blood volume per beat scales with mass (M) and blood flow rate is proportional to metabolic rate (B), heart rate scales as M^(-1/4). This means larger animals have slower heartbeats. The Etruscan shrew has 1200 beats per minute, while an African elephant has 30.

Lifespan is theorized to be inversely proportional to the rate of metabolic damage accumulation (M/B), thus scaling as M^(1/4). This explains why shrews live 1-2 years, while elephants live up to 70.

Crucially, the total number of heartbeats in an animal's life is heart rate multiplied by lifespan. Since heart rate scales as B/M and lifespan as M/B, these terms cancel out, leaving a constant. This explains why nearly all mammals, from shrews to elephants, experience approximately one billion heartbeats in their lifetime.

The Human Anomaly and Urban Scaling

Humans are a significant outlier, experiencing nearly three billion heartbeats in a lifetime. This increase, particularly since the mid-1800s, is attributed to advancements in germ theory, sanitation, and medicine, which drastically reduced child mortality and disease. This demonstrates how science and technology have effectively granted humans an "extra life." Other mammals also live longer in captivity, away from natural hazards.

Interestingly, the trend of increasing human lifespan closely mirrors the growth of urban populations. While this doesn't imply cities cause longer lives, it challenges the historical perception of cities as unhealthy environments.

Geoffrey West and his collaborators have extended scaling laws to cities. They found that certain properties scale superlinearly with population. For example, serious crime scales with an exponent of 1.15, meaning a doubling of population leads to 2.2 times more crime. Wastewater and AIDS cases also follow this superlinear pattern.

Conversely, infrastructure needs scale sublinearly. Gas stations, roads, and electrical cables scale with an exponent of about 0.85. A doubling of population requires only about 74% more gas stations, representing significant savings. This suggests that shared resources make cities surprisingly efficient, even "green."

Even more significantly, total wages, GDP, and patent filings also scale superlinearly, with exponents around 1.15. This means a doubling of city size leads to 120% more of these socioeconomic factors. Comparing a town of 50,000 to a city 100 times larger, infrastructure needs only increase by a factor of 50, while wages, GDP, and inventions increase by a factor of 200. On a per-person basis, city dwellers require half the infrastructure but gain double the socioeconomic benefits.

This suggests cities are powerful drivers of scientific and technological progress. The perception that life in cities feels faster is also supported by data: people literally walk faster in larger cities. While some, like Wolfgang von Goethe in 1825, have expressed concerns about the accelerating pace of life, humanity has historically adapted.

Ongoing Debates and Future Research

While WBE Theory provides a compelling explanation for biological scaling, it is not universally accepted. Other theories predict similar exponents, and some critiques question the data analysis or the inherent noise in biological data. Peter Dodds, for instance, argues that Kleiber's Law itself might not be entirely accurate.

Studies on a wider range of mammals show that while the three-quarters slope fits larger mammals well, smaller mammals appear to align closer to a two-thirds slope. Recent studies on bird metabolism also suggest a two-thirds slope. This raises the possibility that the centuries-old surface law might be correct for some organisms.

The challenge lies in accurately measuring metabolic rates, especially for large animals, which often leads to error bars that encompass both two-thirds and three-quarters exponents. The research community is currently split, with some upholding Kleiber's Law, others favoring two-thirds, and a growing number suspecting there is no single universal scaling exponent across all life. It's possible that larger mammals scale to the three-quarters, while smaller ones scale to the two-thirds.

Despite these debates, the reality of scaling laws is undeniable. Understanding how things scale is crucial, as it reveals efficiencies in larger systems, whether biological or urban. This knowledge can lead to more discoveries, improved living standards, and potentially even more heartbeats in our lifetimes. The next generation of researchers, equipped with better data and analytical tools, may finally resolve these long-standing scientific debates.