When they transferred a piece of skin, the outer epidermis and the underlying dermis, to these materials, they found that when the cells in the dermis were able to pull strongly against the substrate, they tended to bunch up in numerous evenly spaced spots throughout the skin.

These aggregates squeezed adjacent cells in the overlying epidermis, which activated genes in those cells that made them become a feather follicle. The result: an orderly array of follicles, just like in normal chickens.

When the substrate material was too stiff or too soft for cells to push and pull against, the dermal cells all glommed together into one bunch, squeezing the epidermis to yield just one follicle.

Within the ideal range of substrate stiffness, the stiffer materials produced more closely packed follicles, which could explain the fact that feathers are differently spaced on different parts of the chicken.

"The fundamental tension between cells wanting to cluster together and their boundary resisting them is what allows you to create a spaced array of patterns, in an emergent way," Rodrigues said.

The researchers also discovered that squeezing the epidermal cells jostled a protein, beta-catenin, which normally sits on the surface of the epidermal cell, making it relocate to the nucleus. There it flipped a switch that started the genetic program to make each clump of cells differentiate into a feather follicle.

In terms of application, Rodrigues said, "We have gotten really good at taking stem cells and making them progress into different fates, but we are not good at controlling in a clean way how those cells organize into unique structures as they differentiate."

"They need to adopt the right fate, but also the right architecture ... We are presenting design principles that the embryo is using that could be used to rethink the way we do tissue engineering in a dish, specifically thinking about how the mechanics and the cell fate decisions are working together."