3 ms·

I don't think the example of metabolic rate changing with scale is as interesting as Geoffrey apparently thinks it is. Natural selection doesn't produce organi

I don't think the example of metabolic rate changing with scale is as interesting as Geoffrey apparently thinks it is. Natural selection doesn't produce organisms that are "all over the map"-- for example, the mitochondria of a human are not that different from those of a rat, or even an earthworm. Natural selection keeps tweaking the same designs. Often this results in solutions that aren't optimal. For example, maybe a cheetah would be a better animal if it had wheels. But they don't, because evoluation is a kind of hill climbing and You Can't Get From Here to There (tm). Larger organisms don't need to generate as much heat as smaller ones, because heat dissipation occurs across surface area, which increases as the square of mass, whereas heat generation increases as the cube (assuming a spherical cow, but close enough.) This isn't particularly deep or profound, just basic physics. New York isn't a whale, and Microsoft isn't an elephant. This whole thing just reminds me of the endless, pointless debates comparing Linux to a truck, and Windows to a car, that people used ot have on Slashdot years ago.

reality_czech 10y ago
By the way, there is a well-written and thought-provoking essay by Steven Jay Gould about the relationship between the size of animals and their shape. http://hermiene.net/essays-trans/size_and_shape.html http://hermiene.net/essays-trans/size_and_shape.html . He mentions Galileo as the first one to comment on this!
schiffern 10y ago
>Larger organisms don't need to generate as much heat as smaller ones, because heat dissipation occurs across surface area, which increases as the square of mass, whereas heat generation increases as the cube (assuming a spherical cow, but close enough.) Except that hypothesis doesn't predict the correct exponent. If that were the reason metabolic power would scale as mass^(2/3), but it actually scales as mass^(3/4). https://en.wikipedia.org/wiki/Kleiber's_law https://en.wikipedia.org/wiki/Kleiber's_law West's space filling fractal hypothesis predicts dozens of non-obvious scaling laws, not just metabolic power. http://www.youtube.com/watch?v=DFFVSvAr7Wc http://www.youtube.com/watch?v=DFFVSvAr7Wc >the mitochondria of a human are not that different from those of a rat, or even an earthworm. West is acutely aware of this. In the talk I linked above he describes biological lifespans in terms of the relationship between the energy required to synthesize ATP and kT (ie the Boltzmann constant times ambient temperature, being the average energy of a particle due to random thermal motion). The trick is in explaining quantitatively why those universal biological facts scale over size and time with the sometimes bizarre scaling factors observed. It does boil down to physics of course, but there's a lot more to it than just heat dissipation.
- reality_czech 10y ago
  That's a fair point-- a naive hypothesis based only on heat dissipation won't predict the correct exponent. My criticism was basically mostly about the lack of citations in the article to all the earlier people who have thought about these problems, as well as the idea that finding patterns in biological or anthropological data was somehow new and revolutionary. Maybe the article snipped a lot of the context of Dr. West's own writings-- that would be typical for science journalism.

reality_czech 10y ago

By the way, there is a well-written and thought-provoking essay by Steven Jay Gould about the relationship between the size of animals and their shape. http://hermiene.net/essays-trans/size_and_shape.html http://hermiene.net/essays-trans/size_and_shape.html . He mentions Galileo as the first one to comment on this!

schiffern 10y ago

>Larger organisms don't need to generate as much heat as smaller ones, because heat dissipation occurs across surface area, which increases as the square of mass, whereas heat generation increases as the cube (assuming a spherical cow, but close enough.) Except that hypothesis doesn't predict the correct exponent. If that were the reason metabolic power would scale as mass^(2/3), but it actually scales as mass^(3/4). https://en.wikipedia.org/wiki/Kleiber's_law https://en.wikipedia.org/wiki/Kleiber's_law West's space filling fractal hypothesis predicts dozens of non-obvious scaling laws, not just metabolic power. http://www.youtube.com/watch?v=DFFVSvAr7Wc http://www.youtube.com/watch?v=DFFVSvAr7Wc >the mitochondria of a human are not that different from those of a rat, or even an earthworm. West is acutely aware of this. In the talk I linked above he describes biological lifespans in terms of the relationship between the energy required to synthesize ATP and kT (ie the Boltzmann constant times ambient temperature, being the average energy of a particle due to random thermal motion). The trick is in explaining quantitatively why those universal biological facts scale over size and time with the sometimes bizarre scaling factors observed. It does boil down to physics of course, but there's a lot more to it than just heat dissipation.

reality_czech 10y ago
That's a fair point-- a naive hypothesis based only on heat dissipation won't predict the correct exponent. My criticism was basically mostly about the lack of citations in the article to all the earlier people who have thought about these problems, as well as the idea that finding patterns in biological or anthropological data was somehow new and revolutionary. Maybe the article snipped a lot of the context of Dr. West's own writings-- that would be typical for science journalism.