IN 1958 Francis Crick, one of the co-discoverers of the double-helical structure of DNA, spelled out what came to be called the “central dogma” of molecular biology. In a nutshell, this says that DNA makes RNA, which makes proteins. In other words DNA—which carries an organism’s genetic code—“writes” that code into bits of RNA, a similar, but not identical molecule. These then act as messengers which tell a cell’s protein-making machinery what to make.

It is a pithy and memorable summary. Sadly, reality is not quite so clear-cut. In a paper in Science, Sandra Garrett and Joshua Rosenthal of the University of Puerto Rico illustrate how the instructions in the DNA are not always followed faithfully. The RNA message can be rewritten before it is read. And that provides an extra opportunity for evolution to occur.

Dr Rosenthal and Ms Garrett were studying octopuses, looking for differences between those that live in warm, tropical water and those that inhabit the poles. They concentrated on the make-up of the ion channels in the animals’ cell membranes. These channels are cylindrical assemblages of protein molecules which help to control such things as the electrical activity of nerve cells and the release of hormones. The two researchers suspected that the channels found in warm-water species would not work well in the freezing temperatures that their polar cousins endure.

That turned out to be correct. What was odd was that the genes for the proteins involved were almost identical in warm- and cold-water animals. This surprised Dr Rosenthal and Ms Garrett, who had expected that natural selection would have changed the DNA, and thus the composition of the resulting protein.

Instead, differences in composition between warm-water and cold-water ion channels were the result of a phenomenon called RNA editing, in which special enzymes alter the structure of the RNA messenger, and thus of the final protein.

Though RNA editing has been observed before, in animals ranging from humans to nematode worms, this is the first time an edit has been tied to a clear evolutionary difference caused by a feature of the environment—in this case ambient temperature. Of course, it is not strictly a departure from Crick’s dogma. Enzymes, too, are proteins, and so are the transcription factors that regulate their production. Eventually, when the chain of causation is traced in full, the chances are that the underlying difference between polar and tropical octopuses will be in the DNA itself. RNA editing of this sort does, however, provide another way to drive evolution, and may help explain why animals (as opposed to, say, bacteria) are so complex.