Melanoblasts are the precursors to the melanocytes that pigment the skin. Shortly after they differentiate, only a small number exist (around 12), occupying only a small area in the dorsal region of an embryo. By birth, these cells have spread and multiplied, densely populating all parts of the skin. A number of models have been put forward to explain the migration and expansion of melanoblasts, including contact inhibition of locomotion and random migration / proliferation. Another possibility is that melanoblasts are attracted by a ubiquitous attractant that they themselves degrade, thus confining its attractive potential to areas that cells have yet to reach. In this talk I present data from a mixture of skin explants, live whole-embryo imaging and computational modelling to put forward an argument for skin colonisation via two such self-generated attractant-mitogen gradients, one in the dermis, one in the epidermis. I will show that such a model makes excellent whole-organism pattern prediction for several known mutations in genes associated with pigmentation.