Embryos in Deep Time: The Rock Record of Biological Development

7. Fossils and Developmental Genetics

126 Ontogenies do not fossilize. But structures that do were once the result of a developmental process. Fossils of adult individuals can then be informative about development by virtue of preserving phenotypes with an immediate, clear correlation to a specific developmental process. To reconstruct such a process, it is important to consider the position of the fossil in the evolutionary tree of life, to ensure that the analyses are based on correct assumptions. This approach is called extant phylogenetic bracketing. It was introduced specifically to infer soft anatomical properties and behavioral reconstructions in fossils, but it can also be used for reconstructing extinct ontogenies. The extinct animal is compared to its nearest living relatives. One examines the distribution of a developmental feature among extant taxa that “surround” the extinct taxon on a given tree. The feature exists in extant species of a group within which the fossil of interest is placed: one can deduce that it existed in the common ancestor and therefore in the fossil. seven Fossils and Developmental Genetics 128 / Fossils and Developmental Genetics regarding fibrolamellar bone and singular growth curves with rapid phases in some lineages. One may question the scientific value of inferences made about organisms separated by several hundreds of millions of years of independent evolution. But such inferences are justified based on empirical evidence gathered from living organisms. A major discovery of the past few decades is that many developmental mechanisms and molecules are the same across disparate and phylogenetically distant taxa. It is well known that many molecular processes that control phenotypic change are much older than the group expressing those phenotypes, what Sean Carroll has called “ancestral complexity.” This conservatism is valid for very disparate and distantly related species. This leads to the fact that there are deep or fundamental levels, early in geologic and developmental time, at which structures found in the adults of organisms as distinct from one another as flies and humans can be traced back. In this case we talk of “deep homology ” of those structures, meaning “sameness,” or homology, at some level of organization (molecular, cellular) and at some level of the hierarchy in the evolutionary tree of life. Using the present to infer the past, as in the phylogenetic bracket case, may seem limiting. In fact, the evolutionary biologist Mark Pagel has written that this approach “condemns the past to be like the present. Worse, perhaps, the past that we get from looking backwards is a very ordinary past, an average past.” But we have no choice, and it does not mean that we do not find significant and even surprising patterns by looking at fossils this way. And what is “ordinary” or “average” anyway? There are no living trilobites, but there is every reason to assume that the body segmentation they experienced during individual develop- Fossils and Developmental Genetics / 129 ment was associated with the same processes as those recorded in living arthropods. In the following examples I set out to demonstrate how we can learn about developmental evolution by looking at extinct adult phenotypes. verteBral nUmBers Complex animals consist largely of repeated parts, and vertebrates are no exception. This is not obvious at first sight, but if we examine our own or any other vertebrate skeleton, we see the serial repetition of vertebrae diverging only gradually in size and shape. The tissues surrounding these vertebrae are also repeated structures, including muscles, nerves, and vessels. These packages of structures, these segments of the whole, originate during development from building blocks, or somites. Somites form during a particular window of early embryonic development. The rate of segmentation determines, then, how many vertebrae the adult organism will have. Some snakes develop as many as three hundred segments, whereas turtles stop at around twentyfive , with a fixed number of eighteen before the sacrum (the portion of the vertebral column just anterior to the tail bones). There is a segmentation clock, and depending on the species it ticks differently. Sometimes it ticks longer or shorter within species , especially in the tail region. Our own vertebral column is easily subdivided into regions with distinctive features. We have no ribs in the neck, in the cervical region. We have a thoracic region with large ribs, coming together to join a sternum and building a cage where our heart, stomach, and other organs are nicely protected. We have no [18.119.104.238] Project MUSE (2024-04-26 13:41...