screw of a passing steamer. Other errors may result from carelessness, incorrect observation of signs, and neglect to follow the ordinary precautions that should obtain in all researches on identity of the dead body.

Certain circumstances indicative of the mental state of the culprit may throw light on the identity. A person of unsound mind would certainly be suggested as the perpetrator of such a deed as that of the woman already mentioned, who after killing and cutting up her infant, cooked portions of the remains with cabbage and served them at a meal of which she herself partook. Equally conclusive should be the inference in the case cited by Maudsley of a person who, for no ascertainable motive, kills a little girl, mutilates her remains, and carefully records the fact in his note-book, with the remark that the body was hot and good.

The handwriting left by the assassin raight also furnish a strong presumption as to the existence of a mental lesion, since the writing of the insane is often characteristic, especially in the initial stage of dementia. I recall the case of a former patient, an aphasic, imprisoned for having stabbed a man in the abdomen and for having wounded his wife in such a way that her arm had to be amputated. Having lost the power to express himself phonetically, this man used a book and pencil, but his writing showed a degree of agraphia which alone would establish his identity beyond a doubt.

While it is quite possible that dishonest transactions, and even theft, may take place by telephone and the voices of the perpetrators may be unmistakable between distant cities, it is more likely that the phonographic registration of speech or other sound by means of a gramophone should become a matter of medico-legal investigation and a possible means that may lend great assistance in establishing personal identity. Although no precedent may be cited, it is not going into the domain of theoretical hypothesis to mention a discovery of such real scientific certainty that for years after death, and thousands of miles away, gives an indefinite number of reproductions that cannot possibly be mistaken by any one familiar with the voice before it had become "Edisonized." Some gramophone disks lately shown me from Germany registered greetings and messages to relatives in Washington, who were

delighted to recognize the exact reproduction of familiar tones and accents of the Fatherland.

So limitless is the field of research in this direction that there is scarcely an anthropological, biological, or medical discovery that may not sooner or later be applied with profit in the investigations of personal identity where the combined efforts of an attorney and an expert are required.

After the most rigid and scrutinizing anatomical and material examination is made and the closest inquisition entered on, it may often be impossible to give a reasonable explanation for the cause of the physical facts observed. The medical man should remember that his is the one great exception to the rule that rigidly excludes opinions, and that scientific men called as witnesses may not give their opinion as to the general merits of the case, but only as to the facts already proved. This qualifying rule being altogether reversed in investigations into personal identity, and the physician's opinion as to identity being indispensable, it becomes a matter of most serious import that this opinion should be grounded upon absolute and wellattested facts.

MEDICO-LEGAL DETERMINATION

OF

THE TIME OF DEATH.

BY

H. P. LOOMIS, A.M., M.D.,

Professor of Pathology in the University of the City of New York; Visiting Physician and Curator to Bellevue Hospital, New York; Pathologist to the

Board of Health, New York City; President New

York Pathological Society, etc., etc.