Dinosaurs Without Bones

by Anthony J. Martin

  The oldest known dinosaur eggs come from Early Jurassic rocks, or about the same time ichthyosaurs were using live birth. Dinosaurs, however, were very likely laying eggs in the Late Triassic Period, too; but these may have been soft, leathery eggs that were not as easily preserved as the Jurassic ones. The Early Jurassic fossil eggs, which were preserved because of calcite in their shells, give us a minimum time for when dinosaurs had developed mineraized eggs. Basically, this made eggshells that went from “squish” to “crunch” when another dinosaur stepped on them. Hence, the eggs we are so familiar with in our everyday lives, thanks to those most famous of modern dinosaurs (chickens), got their origins relatively early in the evolutionary history of dinosaurs.

  Despite the fact that dinosaur eggs are technically considered body fossils, paleontologists were still motivated enough to devise a classification scheme that helped distinguish different types of eggs. Paleontologists, using a system parallel to a biological classification, came up with ootaxonomy, in which the “oo” prefix refers to eggs. These categories, based on shell microstructure, arrangements of the pores, and overall forms, were given linguistically daunting names such as Spheroolithidae, Ovaloolithidae, Megaloolithidae, Dendroolithidae, Faveoloothidae, and at least a half dozen others. Despite such names causing computer spell-checkers to run and hide, they are essential for the small and dedicated group of dinosaur-egg paleontologists to better communicate with one another.

  Unfortunately, these egg categories do not always exactly match clades of dinosaurs. Although we can be reasonably sure that theropods, sauropods, ornithopods, and ceratopsians produced certain types of eggs or egg clutches, initial identifications are occasionally tested by the discovery of embryonic bones in an egg. Moreover, as of this writing, no one has yet identified an undoubted stegosaur, ankylosaur, nodosaur, or pachycephalosaur egg, which make for surprisingly big holes in our knowledge of the life cycles for these dinosaurs.

  Given that eggs are body fossils, is there any way to elevate their importance further by somehow making trace fossils out of them, too? The answer is yes. At least two species of dinosaurs, Citipati osmolskae and Troodon formosus, provide evidence of post-laying movement of eggs in a nest, an action that must have been done by one or both dinosaur parents. This sort of arrangement constitutes a trace fossil of that parent’s behavior like how a flower arrangement reflects the handiwork of a florist.

  In the case of Troodon nests, distribution patterns and orientations of the eggs provided marvelous clues about female Troodon reproductive anatomy, as well as what she did with the eggs after they exited her body. Paleontologists who studied Troodon egg clutches were surprised to notice that not only were eggs paired but also aligned vertically. Based on their statistically significant pairing, paleontologists surmised that this pattern must have been caused by the mother laying eggs two at a time. Secondly, the eggs were longer than they were wide (elliptical), which meant they naturally should have rolled onto their sides once deposited onto the ground surface. As a result, their vertical orientation means they were righted after laying.

  To best accomplish this feat, the mother or father Troodon likely would have used their hands to turn the eggs upright, then buried the lower end of each egg so that they wouldn’t just roll onto their sides again. No one knows whether Troodon mothers did this with each pair of eggs—two at a time—or whether they waited until the entire clutch was laid before turning them all sunny-side up. In an evolutionary sense, the turning-two-at-a-time scenario seems more likely than the let’s-wait-until-they’re-all-out one. But sometimes evolution works in mysterious ways, going counter to the expectations of us geologically short-lived primates.

  Did dinosaurs ever engage in cuckoo-like behavior, laying their eggs in with the eggs of other species in a nest, and letting some other species’ parents do all of the work of raising the young? Thus far, we only know of one instance in which an egg assemblage contained embryonic bones of more than one species of dinosaur, which was found in Late Cretaceous rocks of Mongolia. The bones showed that one dinosaur was probably Oviraptor (closely related to Citipati) and the other was Byronosaurus, a dinosaur more closely related to Troodon. One would think a discerning mother Oviraptor would have been able to tell whether she was brooding and feeding a Byronosaurus child. But such is the insidious nature of nest parasitism in modern birds: most bird parents don’t know they have a changeling in their midst until it is too late. However, this one instance of possible parasitism is hard to test further, as eggs also could have easily been transported and deposited together in the same place by currents. One way to test this hypothesis, though, would be through trace fossils, such as two distinctive sets of hatchling tracks in a nest. Have hatchling tracks ever been found in a nest? Not exactly, but they have been found very close to nest sites, as will be explained later.

  Another type of interpretable hatchling trace fossil would be an exit hole (“hatching window”) in an eggshell, made by hatchlings as they emerged from their temporary confinement. Sea turtles, for example, have a temporary extension of their beaks when born, which is applied like a can opener from the inside of the egg to open it. Sea-turtle researchers can thus pick up an empty sea turtle egg from a previous year and instantly tell whether its former occupant successfully hatched, died in the egg, or an egg predator got to it before hatching. Similar parts have been described from Late Cretaceous sauropod embryos in Argentina, so these baby dinosaurs could have made such traces in eggs. Have hatchling trace fossils been identified in dinosaur eggs? Yes, which is amazing when one considers how much more fragile an egg would have become once abandoned. Indeed, telling the difference between fractures caused by a hatchling coming out of its egg versus those inflicted afterwards would be very challenging. Nonetheless, trace fossils of “hatching windows” were interpreted in 2002, evident in Cretaceous eggs from Mongolia and China. More such traces could be reasonably proposed for fossil eggs with localized and consistently sized hatchling-appropriate holes in them, especially if more than one is seen in the same clutch of eggs within a nest structure.

  Other trace fossils that might be preserved in eggs would be those from the aforementioned egg predation, where another animal—whether a dinosaur, crocodilian, lizard, snake, mammal, or insect (depending on when in geologic time this might have happened)—nibbled, gnawed, bit into, or chomped an egg. Yet these trace fossils have not been discovered yet either, despite many decades of our demonizing mammals for eating too many dinosaur eggs toward the end of the Mesozoic Era.

  We also currently do not know whether dinosaurs laid unfertilized eggs—like chickens or some other birds—but we presume that if their eggs are found in a nest structure, these were probably fertilized. The extra effort required to build a nest represents an investment of time, resources, and energy and would have exposed parent dinosaurs to predators by keeping them in the same place for a while, thus making them predictable. Instincts that would have impelled parent dinosaurs to make nests with every laying of an unfertilized egg clutch would have been quickly selected out of those lineages.

  An adage I often tell my students when discussing extinction is: “If you want to make a species go extinct, stop it from reproducing.” The huge success of dinosaurs throughout nearly every land environment within much of the Mesozoic Era attests to how they reproduced just fine, probably aided through the vast majority of them making nests for their egg clutches at the right times and in the right places.

  Starting a Dinosaur Family: Building Nests

  Given that at least some, if not most, dinosaur mating successfully resulted in fertilization, and a mother dinosaur started internally producing eggs soon afterwards, the next step for either her or both parents should have been nest building. Although this supposition might be taken for granted, what is surprising to many people is that actual trace fossils of dinosaur nests are only known for a few dinosaurs, and the details of how these were made are still a bit murky. The known nest builders are:


  The ornithopod Maiasaura from the Late Cretaceous in Montana.

  The theropod Troodon, also from the Late Cretaceous in Montana.

  Unidentified therizinosaurs from the Late Cretaceous of Mongolia, which may have even formed nesting colonies.

  Titanosaur sauropods (not yet identified precisely) from the Late Cretaceous in Argentina and Spain.

  The prosauropod Massospondylus from the Early Jurassic in South Africa.

  The ceratopsians Psittacosaurus from the Early Cretaceous in China, and Protoceratops from the Late Cretaceous in Mongolia, although their nests are interpreted more on the basis of many same-sized hatchlings, tightly packed in a small space.

  Other dinosaurs that we are sure made nests, even though we do not have direct evidence of a nest structure, are the Late Cretaceous theropods Oviraptor and Citipati from Mongolia. In a famous case of dinosaur-parenting misattribution, paleontologist Henry Fairfield Osborn, of the American Museum of Natural History, found an Oviraptor skeleton on top of an egg clutch during an expedition to Mongolia in 1923. Nonetheless, he figured the eggs were those of the small ceratopsian Protoceratops, whose bones were abundant in the area, too. In Osborn’s scenario, Oviraptor died and was buried while raiding the nest of another dinosaur, not while taking care of its own eggs. This idea was disproved much later, and through a closely related dinosaur, Citipati. In one of many spectacular finds from American Museum of Natural History expeditions to Mongolia in the 1990s, a nearly complete specimen of Citipati osmolskae—missing only its head—was found in a sitting position above a clutch of long oval eggs. Even better, the clutch was oriented radially. This suggests that the mother or father Citipati arranged the eggs after laying, with such a pattern constituting a trace fossil in itself. When viewed from above, the skeleton and eggs paint a striking portrait. The dinosaur’s arms are held wide in a semi-circle around the eggs, evoking images of a protective mother or father that died in a last vain effort to keep their unborn children from harm. Paleontologists have also documented direct associations of dinosaur parents and eggs—some of which have embryonic bones linking them with their parents—with another specimen of Citipati and several examples of Oviraptor.

  Take away a parent dinosaur’s skeleton, though, and discerning a dinosaur nest becomes a little more challenging. For instance, a cluster of identical dinosaur eggs is sometimes considered as indirect evidence of nests. However, unless these eggs are inside a bowl-like structure, show some sort of post-laying arrangement, or have remains of a parent dinosaur on top of the eggs, a collection of eggs may not be a nest after all.

  This doubt is cast because the eggs may have been moved after laying. As anyone who has done an Easter-egg roll can attest, or had an egg fall from a kitchen counter to the floor while trying to make an omelet, well-rounded eggs can move easily from one place to another under the influence of gravity. In nature, eggs can be rolled or floated by currents, then deposited in low-lying areas with other debris that, to an untrained eye, might look like someone’s idea of a nest. Indeed, the discovery of a dinosaur egg—containing embryonic bones, no less—in Late Cretaceous shallow-marine deposits in Alabama shows that at least one dinosaur egg was capable of floating a long way from land. As we learned previously, dinosaurs may have occasionally swum, but like modern birds or reptiles, were not inclined to make underwater nests.

  In order to better understand what constitutes a dinosaur nest as a trace fossil, a good starting place is to look at the traces associated with modern vertebrates that lay eggs and make nests, such as those of reptiles and birds. If talking just about reptile nests, these can be summarized into two broad categories: ground nests and hole nests, with ground nests made on the ground surface and hole nests below the surface. From there, each type can vary. For example, ground nests can range from simple hastily scraped depressions made by some lizards, to large mounds of vegetation amassed by alligators. Hole nests are also diverse, ranging from the blunt, shallow holes of freshwater turtles and tortoises, to the more elaborate excavations of sea turtles and crocodiles.

  On the other hand, bird nests are as crazily diverse as birds themselves, almost defying facile categorization. Bird nests range from simple scrapes in the ground, to better defined excavations, to tunnels in the ground or in vertical bluffs, to elaborately woven and architecturally complex structures made of a wide variety of natural and man-made materials. Some birds resemble reptiles in their nesting behavior by constructing nests on the ground or as underground burrows, but flight has also made it easier for birds to build nests well above ground surfaces, such as in trees, cliff faces, or buildings. Furthermore, by “in trees,” this is sometimes literal for woodpeckers that actually bore into tree trunks with their beaks, hollowing out areas to lay their eggs and raise hatchlings inside trees.

  How did dinosaurs fit in “reptile vs. bird” models for nesting behavior? It turns out they were somewhere in between, although definitely leaning toward behaviors we observe today in some reptiles and ground-nesting birds. For one thing, every dinosaur nest structure recognized thus far is a ground nest. Hole nests like those of sea turtles or crocodilians, or more fancy structures such as those of some tree-dwelling birds, aren’t yet known for dinosaurs. Of course, this current lack of evidence does not necessarily mean that no dinosaurs made either underground or arboreal nests. For instance, one good candidate for underground nesting would have been the small Cretaceous ornithopod Oryctodromeus cubicularis, which made dens for raising its young (discussed more in the next chapter). For arboreal nests, a few small feathered tree-climbing, gliding, and flying non-avian dinosaurs are known from Early Cretaceous rocks of China. So these dinosaurs feasibly could have made their nests in trees, just like modern birds. However, until we have some evidence for these, we can only speculate.

  For those people who might wonder how to tell a hole in the ground from, well, other things, how would someone recognize a dinosaur nest in the fossil record? Fortunately, paleontologists who have interpreted the few indisputable dinosaur nests in the geologic record made a nice little checklist for the rest of us to follow, which helps considerably. Here is a summary of that list, but posed as questions to ask when encountering a depression in a Mesozoic rock that might be a dinosaur nest:

  Does the depression cut through any layered sedimentary rock below it? This is an especially good question to ask when investigating whether the rock that preserves a potential nest has different colors and grain sizes than the sediment above it. For example, the sediment below the depression might be a green rock mostly made of mud (mudstone), whereas the sediment filling it might be a red rock composed of sand (sandstone).

  Does the depression hold lots of entire (or nearly entire) eggs, along with skeletons of baby or otherwise young dinosaurs, and all of the same species? Be careful with this one, though. As mentioned before, eggs can be transported far away from where a mother dinosaur originally laid her eggs, meaning such a depression might simply be where these eggs and skeletal remains accumulated after being washed about by currents.

  Does the depression also have a raised rim around it, accentuating its basin-like appearance? A related question to ask is, is this rim also made of sediment differing from what is underneath or outside of it? These rims would have served an important primary purpose, such as preventing rounded eggs from rolling away as a mother dinosaur laid them.

  Does the rock filling in the depression and covering the rim have its own distinct traits, like different colors and textures (mentioned before), bedding, or insect burrows and cocoons? Such features might hint of a fossil soil (paleosol), which would tell paleontologists that the depression was exposed at the ground surface before being covered, as opposed to, say, at the bottom of a river.

  However facile this checklist might seem, it represents a lot of previous work on dinosaur nests. Knowledge of some of the history behind it may provide a perspective on how what we know now about dinosaur nests also reminds paleont
ologists how much we still need to learn about these trace fossils.

  The First Recognized Dinosaur Nests

  Anyone who has not studied the history of dinosaur studies might be astonished to know that the first genuine dinosaur nests were not interpreted until 1979. At the time of this discovery, more than a hundred years had elapsed between paleontologists first linking fossil eggs to dinosaurs, which was in 1869. Furthermore, only a few of the previously mentioned criteria were applied to these nests, showing how far paleontologists have come since then in defining them.

  In what became a revolutionary discovery, inspiring nearly everyone to reconsider what they thought they knew about dinosaur parenting, John (“Jack”) Horner and his friend Robert (“Bob”) Makela, while prospecting Late Cretaceous rocks in Montana, found depressions filled with eggs and partly grown juveniles of the large ornithopod dinosaur Maiasaura peeblesorum. Horner and Makela inferred the presence of nests on the basis of many nearly entire and identical eggs, some containing embryonic skeletons of Maiasaura, which were in the bottom of what looked like indentations to them. These former hollows were visible as differently colored sediment above and below the eggs; unfortunately, though, no sedimentary rims were present. So either these dinosaurs did not make such rims, the rims had been eroded and not preserved, or these researchers missed them because of too-subtle differences in the sediment.

  Still, on the basis of localized assemblages of eggs and baby dinosaur bones in the same small areas, they hypothesized that this part of Montana was a Maiasaura nesting ground about 75 to 80 mya. The close spacing of the nests also hinted, for the first time, that dinosaurs nested communally. Subsequent field work done in the same area yielded more and different eggs, which at first were linked to the small ornithopod dinosaur Orodromeus makelai (yes, this species was named in honor of Makela), although later they were connected to Troodon formosus. Hence, for the first time paleontologists began thinking of dinosaurs and their nesting behaviors as more akin to those of ground-nesting birds and not reptiles. Even more exciting, the find of one nest with fifteen juvenile Maiasaura in it, all the same size and apparently in the same age range, was strong evidence favoring extended parental care in this species of dinosaur. Otherwise, why stay in a nest unless your parents are feeding you? (All of you parents with children who have graduated from college yet are still living at home, you may now wearily nod your heads in agreement.)