How was the weather in the Jurassic period? Amazingly, even though it was almost two hundred million years ago, we can say that the Jurassic was warm and wet. The lush vegetation produced major coal deposits, and life thrived.

The reason we know this is that certain geological formations – rocks, fossils and ancient ice sheets – preserve records of temperatures from the time they were formed. This ability to infer ancient climates not only enables us to reconstruct a history of the world before humans were a glimmer in some reptilian eye, but also gives us a vital context for evaluating changes in global climate today. For instance, the planet is warmer on average now than it has been for millennia, and we know that such balmy times have occurred before over the past many millions of years, typically coinciding with periods when the air was full of greenhouse gases released by intense volcanic activity.

Current methods for reconstructing the world’s temperature in the past – so-called palaeothermometers – each have their own strengths and limitations, which is why new ones are always welcomed. A team at the University of Wisconsin, led by physicist Pupa Gilbert, has now proposed one, which depends on looking at mollusc shells under the microscope.

If that seems an unlikely place to find a thermometer, consider some of the existing ways of taking the prehistoric temperature.

Taking temperature

One way is to measure different forms of the same chemical element, or isotopes, in the shells of marine organisms, which get deposited as sedimentary rocks. For instance, the relative concentrations of two common isotopes of oxygen, called oxygen-16 and oxygen-18, are often used to deduce the temperature of the sea surface at the time the shells were formed. This is because water molecules containing the two different isotopes have different rates of evaporation – so the higher the temperature, the more evaporation and the bigger the difference in amount of the two isotopes left in the sea water. There is a complementary record of oxygen isotopes in the water that fell as snow to form the great ice sheets at the poles.

Another temperature record in the chemical composition of marine sediments is the ratio of calcium carbonate to magnesium, which reflects the ocean temperature when the sediments were deposited. And a region’s historical temperature also affects the growth rate of trees, as recorded in tree-ring widths, and the kinds of fossil plants and animals found there.

All of these methods, however, have complicating factors that can blur the temperature record. And they apply only over particular timescales – the lowermost ice in the ice sheets, for example, is only a few hundred thousand years old, while tree ring records rarely go back even as far as the last ice age. In contrast, say Gilbert and colleagues, molluscs have been making shells from a substance called nacre since the Ordovician period 450 million years ago, and this hard mineral is abundant in fossils – in ammonites, for example.

Crystal maze

Nacre is remarkable stuff. Commonly known as mother of pearl, it is basically calcium carbonate, but the organisms build this protective shell around their fleshy parts in the form of flat platelets stacked into layers. The brickwork stacking helps to make the shell not only hard but also very strong. While a solid block of the mineral could fracture like brittle crockery, the sheet-like array stops a crack from advancing by deflecting it sideways, prising the layers apart and exhausting the crack’s energy in the process.

By looking at how electron and X-ray beams are absorbed by and bounce off samples of nacre, Gilbert and colleagues have found that the precise microscopic structure of these layers – the width and thickness of each platelet, and the order of the nacre crystal orientations – varies from one species to another. In particular, there was more disorder in the crystal lattice orientations for species that live in warmer waters. And for at least one of the organisms studied, a pearl oyster from French Polynesia, these variations in orientation at different depths within the shell wall correlated strongly with the water temperature in which that part of the shell was formed. As well as this, the thickness of the platelets seems related to the water pressure, and thus the depth, at which the organism grows.

If the researchers can firmly establish that temperature really does directly determine whether the nacre platelets grow with their crystal lattices more or less aligned or oriented more randomly, while pressure determines the platelet thickness, then ammonites and their ilk could offer a valuable way of figuring out not only what the climate was like hundreds of millions of years ago but also how the temperatures of the seas varied with depth. That could supply a three-dimensional temperature map of the oceans within which the first fish evolved, and from which they eventually crawled out onto land.

If you would like to comment on this story or anything else you have seen on Future, head over to our Facebook page or message us on Twitter.