years ago, the same time, it turns out, that oxygen levels began to plummet. The immediate successors in terrestrial dominance of the dinocephalians, the dicynodonts, were the dominant herbivores of the time from 260 million to 250 million years ago. They in turn were almost eliminated from the planet in the Permian extinction, which will be described in more detail in the next chapter. The dicynodonts were hunted by three groups of carnivores, all therapsids: the gorgonopsians, which died out at the end of the Permian; the slightly more diverse therocephalians; and the cynodonts, which ultimately evolved into mammals during the Triassic. We will return to these groups—and their horrific fate—in more detail in the next chapter.
The rise of atmospheric oxygen to unprecedented values of over 30 percent in the Carboniferous-early Permian was accompanied by the evolution of insects of unprecedented size. The giant dragonflies and others of the late Carboniferous through the early Permian were the largest insects in Earth’s history. Perhaps it is just coincidence but most specialists agree that the high oxygen would have enabled insects to grow larger, since the insect’s respiratory system requires diffusion of oxygen through tubes into the interior of the body and in times of higher oxygen, more of this vital gas could penetrate into ever-larger-bodied insects. So if insects got larger as oxygen raised, what about vertebrates? New data acquired by French anatomist Michel Laurin can be used to test this question. Indeed, it does seem that body size in various late Paleozoic reptilian lineages do track oxygen levels.
Laurin’s published data were compared to the oxygen levels given in the Berner curves (like so much discovered while writing this book, this information is simultaneously being published in the scientific literature). Laurin looked at a sample of animals from the anapsids and synapsids. He used a measure of body length and a measure of skull size to evaluate animal size, for specific time horizons between the late Mississippian and the end of the Permian, from about 320 million years ago until about 250 million years ago. Then I did a simple regression analysis of mean size and oxygen levels. Indeed, mean skull size closely