The paper describing this extraordinary artifact, which is due to be published in the journal NeuroImage next month, reads somewhat like a medieval murder mystery, with elements of archaeology, forensic science and neuropathology. The child's skeletonised remains were exhumed in 1998 from a burial site in the city of Quimper, north-western France, in a wooden coffin with the head wrapped in leather and resting on a pillow. The coffin was dated by dendrochronology (tree-ring dating) to the mid- to late 13th century (1250-1275 AD), and the age of the child determined by examination of the teeth. After the remains were discovered, the brain was removed and immersed in a preservative formalin solution. Several years later, Christina Papageorgopolou of the University of Zurich's Institute of Anatomy and her colleagues began to re-examine it
Link - via techmedicus
From the Upcoming ueue, submitted by Geekazoid.
Personally, I think that's the degree of removal I'd be okay with as a rule of thumb, though I'm sure I could come up with exceptions in either direction.