This project will prepare three dimensional digitized images of fossil endocranial casts (endocast = "brain" cast) for a databank to be available on the internet on brain evolution. These will become fundamental data on the fossil evidence on the evolution of the brain. The photograph of Bathygenys reevesi (click on "Back" to see it again) was the result of a 2-D scan. It is adequate for measurements of length and width, but for information about surface area, one requires 3-D scans. We are interested in measuring surface area, because in living mammals, surface area of the cortex of the brain is closely related to the brain's information-processing capacity. Following the "uniformitarian" scientific principle, we assume that this was also true for fossil mammals. We can, therefore, learn more about the evolution of information-processing capacity from 3-D images, and we now also have software available to make those measurements. This was almost impossible to do with irregular solids before the advent of computers.

To prepare data for this project, one scans a fossil, such as the endocast of Bathygenys reevesi, and records x-y-z coordinates of the surface. A preferred method is to scan with a laser camera system, such as that provided by Cyberware and headus. You can examine true 3-D images created by their Model 15 Scanner of Merycoidodon culbertsoni. This oreodont lived some 5 million years after Bathygenys. Its endocast, which is missing olfactory bulbs and parts of the hindbrain, is 77 mm long and 45 mm wide at its widest part (cerebral hemispheres). The shaded region of the surface, shown below the set of three copies, is prepared for surface measurement, which is done by appropriate software. The software, thus, dissects the "surface" to provide partial surface areas or volumes associated with those surfaces. For example, one could differentiate mammalian fossil brains into "virtual" fractions representing neocortex, various gyri, paleocortex, cerebellum, olfactory bulbs, and medulla, as well as cranial nerves, if the endocast is good enough (many are). The "dissection" can be unilateral or bilateral. The surface can also be "rendered" to recover a realistic image of the fossil.

The scientific justification and utility of such data is indicated in two publications (Jerison, 1990, 1991a) in which 2-D projections of the lateral surface of the neocortex are analyzed in a sample of 35 fossil and 24 recent species of carnivores and ungulates. Those publications present the only unequivocal evidence for a "progressive" increase in the amount of neocortex ("neocorticalization") during the past 60 million years. The results were preliminary, because (as mentioned in those publications) 2-D images provide only suggestive and not definitive data on the areal extent of the brain structures that can be studied. The 2-D data were collected in 1987 with a planimetric method, when there were formidable technical problems in measuring surface of irregular solids such as brain casts. These were only beginning to be solved at the time and required mainframe computers. They are presently solvable with commercially available PC software and hardware as described above. Endocasts and bones can be scanned for the creation of manageable files containing the necessary data.

More information on the topic is in Jerison (1991b, 1994). Related data are available in Finlay & Darlington (1995), to analyze the evolution of functional as well as structural neural systems.

---