THE PERCENTAGE OF CASES RESULTING IN A RULING OF 'UNDETERMINED CAUSE OF DEATH' OR 'UNDETERMINED MANNER OF DEATH' WAS 1.73 PERCENT, WHICH IS AT THE LOWER LIMITS OF THE FEW OTHER SERIES QUOTED IN THE LITERATURE. BECAUSE THE CORONER'S OPINION OF THE CAUSE AND MANNER OF DEATH IS LEGALLY ADMISSIBLE EVIDENCE AND IS TREATED AS FACT, NOT OPINION, MANY MEDICAL EXAMINERS ARE PARTICULARLY CAREFUL WHEN THERE ARE ANY DOUBTFUL CIRCUMSTANCES. IN THIS SERIES THE TYPE OF CASE MOST OFTEN RULED UNDETERMINED WAS A DEATH RESULTING FROM KNOWN TRAUMA BUT WITHOUT SUFFICIENT RELIABLE INVESTIGATIVE EVIDENCE, ON WHICH TO BASE A RULING, OF MANNER OF DEATH (35 PERCENT). THESE CASES INCLUDED FOUR WHICH PROBABLY INVOLVED BATTERED CHILDREN, TWO IN WHICH PRIME SUSPECTS 'PASSED' A POLYGRAPH EXAMINATION, TWO IN WHICH A POLYGRAPH EXAMINATION WAS REFUSED, AND TWO ABORTION DEATHS. THE SECOND LARGEST CATEGORY (25 PERCENT) INCLUDED DRUG-RELATED DEATHS, USUALLY FROM BARBITURATES. IN THE THIRD CATEGORY (22 PERCENT) THERE WAS INADEQUATE HISTORY, PATHOLOGICAL EVIDENCE, OR BOTH ON WHICH TO BASE DEFINITE RULINGS. IN NINE CASES (10 PERCENT) THE BODY WAS SO DECOMPOSED AND THE HISTORY SO INCOMPLETE THAT NO DEFINITIVE RULING COULD BE MADE. DURING THE EARLY YEARS OF THE STUDY THE EXPERIENCE AND TRAINING OF THE PHYSICIAN SEEMED RELATED TO THE NUMBER OF UNDETERMINED FINDINGS, BUT THIS WAS NOT TRUE AFTER 1974. TABLES PRESENT STUDY DATA; REFERENCES ARE APPENDED. (GLR)