Number density of insulin crystals versus nucleation time dependences were measured simultaneously, during the same experiment, at four typical places: in solution bulk, at the glass support, at the air/solution interface, and at the solution/glass/air boundary. Stationary nucleation rates were determined from the linear parts of the corresponding plots, and energy barriers for nucleus formation and nucleus sizes were estimated. A key finding of the present investigation was that, surprisingly, the lowest energy barrier (3.8 × 10−13 erg), and correspondingly the smallest nucleus size (six insulin molecules), were calculated not for some kind of heterogeneous substrate, but for insulin crystals nucleated in the solution bulk; in both cases, the critical nuclei were formed from preliminary built Zn-insulin hexamers. The interpretation of these results is that no true homogeneous but rather heterogeneous (surface) insulin crystal nucleation is taking place also in the bulk solution. The nuclei form on some foreign particles of biological origin that are present in every protein solution.