It is an experiment you could do in a school chemistry lab. But it produces weird growths that, although made purely from inorganic materials, share some of the characteristics of living organisms.

Most chemical mixtures quickly settle into an unchanging state. So the fact that dynamic cell-like structures can arise spontaneously from a simple mixture is a surprise, says Jerzy Maselko of the University of Alaska in Anchorage and Peter Strizhak of the Institute of Physical Chemistry in Kiev, Ukraine, who made the discovery1.

Understanding how this happens could give us clues about how life may have arisen on Earth, or even other planets, where the blends of chemicals present might be quite different from that of the early Earth.

Full flow

Maselko and Strizhak mixed calcium chloride, sodium carbonate, copper chloride, sodium iodide, hydrogen peroxide and starch (see box). They found a fungus-like, soft membrane grows out of the mixture, enclosing a hollow cavity up to 1 cm across. Chemicals diffuse through this membrane, react inside the cavity, and then diffuse out, creating swirling clouds of violet liquid in the green base solution.

Rather than reaching equilibrium, this process persists. The reactions, say the researchers, are reminiscent of the way living cells sustain themselves, driven from equilibrium by the flow of chemicals and energy across their membranes.

Maselko and Strizhak even saw a kind of replication in their chemical brew. Sometimes the cell structures grew into forms with several lobes, or sprouted buds that split off from the parent membrane.

But although they look impressive, can these structures tell us anything about the origin of true life-forms? It seems the answer might be yes, because the differences between the two processes are not as fundamental as one might assume.

Graham Cairns-Smith, a chemist at the University of Glasgow in Scotland, has speculated for many years that life on our planet may not have started with organic (carbon-based) molecules. He suggests life may have begun with inorganic ingredients, such as clay minerals that can carry heritable information in the stacking sequence of their sheets of atoms. Such 'clay organisms' might be able to replicate, Cairns-Smith argues.

Maselko is keen to follow up his discovery to see just how far the parallels with life run. "This is only the beginning," he says. "We will see many other systems like this. The next step will be to get these systems to evolve."