a function of their greater age at death. The average age at death in all II females is 67.2; the average duration 8.8. The average age at death in all 12 males is 62.3; the average duration 2.8. The differences in age at onset: female average 58.4 (atrophics, 59.8), male average 59.5 (atrophics, 62.9), are not great.

5. A comparative study of the distribution and extent in several cortical areas of certain pigmented materials demonstrable by iron-hematoxylin (among other methods) brings out extreme and interesting variations in the cases examined.

Perivascular cell pigmentation is almost uniform in different areas of the same case, bar focal destructive lesions, but varies in degree in different cases.

Neuroglia cell pigmentation, when of general distribution, probably varies more or less directly with age.

Nerve cell pigmentation (iron-hematoxylin) is not a function of age. It is premature to relate the amounts and distributions of nerve cell pigments with different mental diseases.

THE PROPER SIZE OF HOSPITALS FOR THE INSANE.

BY CHARLES W. PILGRIM, M. D.,

Medical Superintendent, Hudson River State Hospital, Poughkeepsie, N. Y.

The question as to the proper size of a hospital for the insane, for the attainment of the best results in care and treatment, is one which has long been discussed.

On June 18, 1850, a committee of the Association of Medical Superintendents of American Institutions for the Insane was appointed and instructed to report at the meeting in 1851 a series of propositions relative to the structure and arrangements of institutions for the insane which would express the well-ascertained views of that body in reference to points upon which there was unanimity of opinion. Twenty-six propositions were reported, upon which all were agreed, and many of them are as pertinent to-day as they were nearly sixty years ago. I will not detain you with all of these propositions, although they are well worth reading, but will refer only to No. V, which says: "The highest number that can with propriety be treated in one institution is 250, while 200 is a preferable maximum," and No. VII, which says: "Every hospital having provision for 200 or more patients, should have at least eight distinct wards for each sex."

For more than 15 years these propositions remained unchanged and during that time the small institution with its small wards, and preponderance of single rooms, remained the standard for asylum construction in this country. In 1866, owing to the difficulty of providing for the increasing number of insane, and after an acrimonious debate by members of the Association in regard to the care of the chronic insane in separate institutions, which was caused by the passage of the act authorizing the establishment of the Willard Asylum, Dr. Nichols offered the following resolution: "The enlargement of an institution for the insane, which, in the extent and character of the district in which it is situated, is conveniently accessible to all the people of such district, may be properly carried to the extent of accommodating 600

patients, embracing the usual proportions of curable and incurable insane in a particular community."

66

This resolution did not stop the building of the Willard Asylum but it did for more than two decades limit, to 600 beds, the size of all institutions which were supposed to treat and cure the insane. There was no great increase in the size of any of the existing institutions in the State of New York, with the exception of Willard, which was established for the care of the chronic insane only, and the asylums controlled by the counties of New York and Kings, which cared for the insane of the Metropolitan district, until the passage of the State Care Act in 1889. With the passage of that act, however, the demand for immediate State provision for those who were being cared for in the county-houses became so urgent that there seemed nothing left but to enlarge as speedily as possible the existing institutions, which soon became the vast caravansaries," as they have been aptly called, from which we now suffer. America, in its hospitals for the insane, as in everything else, took readily to the idea of "bigness" and many of our hospitals exceed in size anything to be found in any other country of the world. Owing to the lesser cost of this style of construction, on account of the large wards and dormitories. and associated dining rooms, this type of building rapidly gained favor, in the eyes of those who considered only the initial cost; but a reaction came as soon as experience demonstrated the disadvantages of this plan and to-day the majority of thoughful and experienced alienists believe that neither the interests of the State nor those of the patient are best served by the present methods of herding the insane in vast wards and dormitories with from 50 to 150 patients in each. In England, where huge institutions at first found great favor, a reaction has also come. Dr. Richard Greene, late superintendent of the Berry Wood Asylum, Northampton, in a thoughtful paper on "The Insane, the Asylum and the Nurse," says: "I need hardly point out that the mere size of an asylum has much to do with the amount of the maintenancerate, and it would seem that the very large and the very small are the highest. Putting aside the very small ones and dividing the rest into two classes, one between 600 and 1200 beds and the other from 1200 to the highest limit yet attained, say 2400, we

should find that in the first section the average maintenance-rate was 8 s. 4 d. and in the other section it was 9 s. 2 d. a week. As there are 33,000 in the latter section, a simple calculation will show that there is a wholly unnecessary loss to the nation of £95,000 a year. I have emphasized the money side of the question, but waste of money is not the only objection to these gigantic asylums, and if any of my audience here to-night possess influence with the Home Secretary, I hope they will entreat him not to sanction the erection or extension of any asylum beyond 1000 beds or at the utmost 1200."

While an analysis of the maintenance-rates in the various hospitals of New York State does not support Dr. Greene's contention altogether, it does lead to some very interesting conclusions. I have carefully examined the figures relating to the net per capita cost of support in the different institutions for the fiscal year ending September 1, 1907, and I find that they make an excellent showing for the smaller hospitals. If we take the four largest hospitals in the State with a total population of 13,699 patients and an average population of 3425 for each institution, we find that the average per capita cost is $170.13 per year, while with four of the small institutions with a total population of 4702 or an average of 1175 for each, we find that the average per capita cost is $169.95 per year. A still better showing is made if we take the lowest net maintenance-rate in any of the smaller hospitals, which is $148.40, and compare it with the lowest net rate in any of the larger hospitals, which is $157.54, for here we find a aifference of $9.14 per capita per year in favor of the smaller institution.

These net figures are obtained by deducting from the gross cost of maintenance the amount collected by each hospital for the support of those patients whose friends are able to fully or partially reimburse the State and, therefore, should not be used without qualification, as the location of the hospitals and the social grade of the patients have more to do with this question than numbers. They are presented merely to show the fallacy of using numbers alone and that by one process of reasoning at least the small hospitals support their patients at a lower per capita cost to the State than the large ones.

Now if we take the gross cost of maintenance in the four smaller hospitals, with an average population of 1175, we find that the average cost is $191.35 per year; the three hospitals with a population ranging between 1500 and 2000 have an average rate of $190.88; those with a population ranging between 2000 and 3000, $191.48, while the four largest with an average production of 3425 have an average rate of $178.93. We thus see that a population of about 1500 gives as great economy as it is possible to get until we extend the institution to twice that number.

If we take all the institutions with a population of less than 1500 we get an average yearly per capita cost, as before stated, of $191.35, while if we take all of the institutions with a population of more than 1500 we get a general average of $183.48, a difference in favor of the large institutions of a little less than $8.00 per capita per year. If we try to discover where this difference is we find that the average per capita cost for medical service in the four smaller hospitals is $13.42 while in the four larger hospitals it is only $8.00 per year; the average cost of the ward service in the former is $36.54 while in the latter it is $33.93. We thus find that the two items of medical service and nursing show a difference sufficient to account for the whole saving. And where is there a medical man who is willing to say that all economies should be effected in medical care and nursing unless he is willing to see our institutions lose their hospital character altogether and become vast boarding houses instead of the curative establishments which we should strive to make them?

In New York State where the financial management of the hospitals is practically on a uniform basis, and where, on account of the system provided for the "joint purchase" of supplies, the small hospital is placed on an equal footing with the large one, there would seem to be but little reason for mammoth institutions. When we analyze the results of treatment we find that the percentage of cures, computed upon the original commitments, was 26.78 per cent in the four smaller hospitals, while 21.26 per cent were discharged as improved, making a total of 48.04 per cent who were restored to home and friends. In the four largest hospitals the percentage of cures was 24.45 per cent, while 20.08 per cent were discharged as improved, making a total of 44.53 per