4 ms·

> "given our understanding of star formation it is difficult to form isolated objects that small." No, that is demonstrably false. You can try this with any pa

• antognini 12y ago
  The problem is that you cannot treat star formation as a set of collisionless particles. In general you can estimate the scale of star formation through the Jeans mass, which is the mass of a gas cloud at which the dynamical time becomes equal to the sound crossing time. This is essentially the scale at which the cloud becomes unstable to perturbations --- pressure is no longer able to support the cloud against gravitational oscillations. The Jeans mass depends inversely on the square root of the density of the cloud. The main problem in forming very low mass objects is that you need very large densities. For a Jupiter mass object you need a density of ~5 x 10^4 / cm^3. Even in molecular clouds it is difficult to achieve densities that large. There are ways around it --- turbulence can produce small regions of very high density, for instance. But like I said, they're not well understood. In general it's much easier to create a low mass object in the vicinity of something bigger because then it can form by fragmenting out of the disk that surrounds the larger protostar.