As mental disease is generally of longer duration than other diseases, especially as seen in asylums for the insane—as it may be of life-long duration, and may frequently recur, or threaten to recur, in a single individual—to obtain a perfect knowledge of insanity, its history should extend from the earliest symptoms, and its subject be traced throughout life. Hence statistics which are made from records extending over a greater number of years are much more valuable than those which embody the results of a shorter period. A perfect series of statistical tables would require to exhibit, as far as they can, the complete history of every case, from the commencement of the mental disease until the death of the individual. This, however, is almost impossible. Dr. Thurnam,∗ under peculiar circumstances, has been able to give a series of statistics which trace the history of individuals beyond the asylum walls; but in ordinary cases such completeness is not obtainable, and we must content ourselves with less perfect results. Generally the history of the patient ceases when he leaves the asylum, and any facts we may obtain in his subsequent career are so few, and possibly inaccurate, that they cannot be embodied as trustworthy statistics. The history of those dying in the asylum, and of those under observation for five years or more, is more complete; but such cases are only those in whom treatment has been unsuccessful. In the cases of readmission, however, we have generally more trustworthy data, and are thus enabled to trace many subjects of mental disease throughout life.