High-tech images shrink fossil braincase.

Current theories about brain evolution in humans and prehistoric hominids may need revision, according to a new study based on computer-generated images of the inside of a fossil cranium.

Computerized reconstructions of a 2.8-to 2.6-million-year-old hominid braincase found in South Africa and attributed to Australopithecus africanus place its volume at about 515 cubic centimeters--much less than an earlier rough estimate of at least 600 cubic centimeters, reports a team directed by anthropologist Glenn C. Conroy of Washington University School of Medicine in St. Louis.

Cranial volume of other australopithecines may have been overestimated in previous studies as well, Conroy's group suggests in the June 12 Science.

"The implication of [the South African specimen's] surprising cranial capacity is that something is very wrong with the published record of early hominid cranial capacities," writes anthropologist Dean Falk of the State University of New York at Albany in an accompanying comment.

Conroy and his coworkers studied a presumably male specimen, discovered in 1989, that retains much of its face and the left side of its cranium. Using anatomical landmarks on remaining parts of the braincase, the researchers generated a complete, three-dimensional computerized tomography (CT) image of the cranium. They then used the computer to create an image of the cranium's inner surface and calculated a volume of 513 cubic centimeters.

They revised that figure to 515 cubic centimeters, based on the overall volume determined by 106 two-dimensional CT "slices" taken throughout the braincase and on the volume of water needed to fill a cast of the fossil cranium.

Preliminary water-volume results for another A. africanus cranium reach only about 370 cubic centimeters, down from a prior estimate of 428 cubic centimeters based on a plaster cast, the scientists add.

During the 1980s, debate raged over how to make and interpret plaster molds of the inner surfaces of fossil craniums. This controversy will not go away, says anthropologist Ralph L. Holloway of Columbia University, because CT re-creations of partial skulls still hinge on researchers' judgments about the presence and location of various anatomical landmarks.

Holloway, who conducted early analyses of cranial capacity in several australopithecine skulls, doubts that CT imaging will substantially alter his original estimates.