their efficiency of oxygen uptake through the evolution of endothermy and by changes in the nasal area of the skull. Soft-part and physiological responses may have taken place as well, such as increasing the number of red blood cells in the body by increasing red-blood-cell-forming bone marrow. The most obvious change would have been in the lungs and circulatory system, and evolution of the four-chambered heart probably happened at this time. But in how many lineages?
Both the avian and mammalian hearts have four chambers: two auricles and two ventricles. Many other land tetrapod lineages, including many reptiles and all amphibians, use a three-chambered heart. This difference affects the relationship between oxygenated blood coming back to the heart from the lungs and venous blood returning to the heart from elsewhere in the body. The latter is depleted of its oxygen content. In the four-chambered system, there is never mixing of these two blood groups. But in the three-chambered heart, mixing can take place, and this would seem to reduce the efficiency and oxygen-carrying capacity of this kind of respiratory system.
Because the four-chambered heart is associated with endotherms of high metabolic activity, it has long been argued that a perfect separation of bloods was selected for in animals that had higher metabolic activity levels, such as endotherms. Recently, this traditional view has been cleverly questioned by a group of Australian biologists led by Roger Seymour. They have pointed out that some snakes are capable of keeping oxygenated and nonoxygenated bloods separated even though they have a three-chambered heart. Instead, they argue, the four-chambered heart evolved for allowing elevated blood pressure rather than blood itself. Four-chambered hearts are larger than the hearts of ectotherms, and high blood pressure seems to be associated with endothermy.
Again, it must be assumed that this (and all other work about respiration discussed to date) was argued in the context of present-day oxygen levels, since this is never brought up. Seymour and his group pointed out that the activity levels of endotherms require more oxygen getting to various parts of the body faster than in ectotherms. Perhaps