Growth law vs shape and topology

Weaire Phelan - large dodecahedra shrink and small "barrels" grow -> does it return to "standard" WP?

Does the same apply to a polydisperse Kelvin foam?

What effect does shear have on the growth rate (G)? on pressures?

Different 3D growth laws predict different values of F at which G=0: how wide is this dispersion? Depends on mu_2? If the dispersion is large, then topology is a bad predictor.

There is a problem at low F - due to liquid fraction?

Shuld we include more topological details? geometrical details? correlations? For the second option, how should we characterise shape? One possiblity is Kraynik's "Q" tensor (bubble stress), but there is no good correlation with G. So what else? Does it make sense to characterise a _group_ of bubbles?

Macpherson-Srolovitz (Nature) "rephrased" the growth law in terms of integral geometry, but no new insight? No easier to calculate mean width than sum(pressures*areas). It gives a difference of two large numbers. (Could it be useful for tomographic images, with only PBS? no, because it is the curvature of the faces that's important.

What if dV/dt ~ edge length? Does this give a model of something? Probably not practical!