What Can Eggshell Tell You? A Fossil Named Without Knowing Who Laid It

Eggshell is a mineral tissue built in layers by the female reproductive tract, and the arrangement of crystals within it is specific to the group producing it, which makes a fragment a few millimetres across identifiable to a family under a microscope even when nothing else survives. Thin sections cut across the shell reveal the layering, the shape of the crystal units, the boundaries between them and the pattern of pores running from the inside to the outside. Dinosaur shell is distinguishable from bird, crocodile, turtle and lizard shell on those features, and within the dinosaurs several distinct structural types are recognised and correspond roughly to major groups. The outer surface carries ornament, ranging from smooth to ridged to nodular, which varies between types and is visible without a microscope on a good fragment.

The channels through the shell carry more information than any other feature:

• •Gas exchange happens through them, since an embryo consumes oxygen and produces carbon dioxide and water vapour
• •The total pore area sets how fast that exchange can happen, which must match the needs of the embryo inside
• •Eggs buried in sediment or vegetation need high conductance, since the surrounding medium is humid and gas moves slowly through it
• •Eggs incubated in open air need low conductance, since high porosity would lose water and dry the embryo
• •Measuring porosity therefore indicates the incubation method without any other evidence
• •Results across dinosaur groups indicate buried nests in some and open brooded nests in others, matching the body fossil evidence where both are available

Because a shell fragment rarely comes with an embryo, the material is classified in a separate system with its own names, structured in the same way as the naming of animals but applied to egg types rather than to species. That parallel arrangement is the same solution used for trace fossils and for the same reason, since naming what is observed avoids asserting an attribution that the evidence does not support. Connecting an egg type to an animal requires either an embryo inside, which is rare and decisive, or an adult preserved on a clutch, which is rarer. A handful of such connections exist and each anchors a whole egg type to a group. The system is occasionally criticised as unnecessary duplication and is defended on the grounds that the alternative would be either silence or guesswork.

Eggshell has been drawn into the argument about what ended the dinosaurs, and the episode is a useful caution. A body of work in the mid twentieth century reported that shell from the final stage of the Cretaceous in southern Europe and India was abnormally thin and showed pathological structure, and proposed that reproductive failure caused by environmental stress contributed to the extinction. The claim was influential and has been substantially undercut, since the samples were small, since shell thickness varies enormously between species and within a clutch, and since abnormal shell occurs in modern populations at low rates without indicating anything about a species' prospects. Later work found no consistent thinning trend. The general lesson is that a measurable quantity can be sampled and analysed correctly and still support a conclusion it cannot bear, if the baseline variation was never established.

Accumulated eggshell evidence has supported several conclusions about how these animals bred. Egg size relative to adult size was small in the large dinosaurs, far smaller proportionally than in birds, which constrains how much the hatchlings could be provisioned and implies rapid growth after hatching. Shell thickness scales with egg size and is limited at the top end, since a shell thick enough to support a very large egg would be too thick for the hatchling to break, which may be part of why dinosaur eggs never became proportionally large. Clutch arrangements recovered in place show eggs laid in pairs in some groups, which indicates two functioning oviducts and distinguishes them from birds, which retain only one. And the distribution of egg types through time records changes in nesting strategy across the group.

Crystal structure and layering identify a fragment to a family under a microscope, and pore density indicates whether the eggs were buried or brooded in open air, which no other evidence supplies. Because attribution to a species is almost never possible, egg types are named in a parallel system. Eggs laid in pairs indicate two working oviducts, unlike birds, which keep one.