would in part help the town clerk to decipher the eligibility of the writing on some of the certificates returned.

Discussion by Jasper P. Newton, Benson, Vermont.

There have been some rather loose statements made in the certificates of death. I had a certificate the other day: age 80 years; old age contributing cause; angina pectoris immediate cause. If a man has angina pectoris he must have died from angina pectoris.

I would like to ask Mr. Wilbur if the terms chief and contributing causes of death could not be replaced by direct and remote?

Discussion by A. A. Butterfield.

To show just how valuable these conclusions are, I cite a case where the cause of death was given as heart disease and the contributing cause as sudden death.

Discussion by Jasper P. Newton.

Has the health officer any duty to perform-is he to amend these certificates of death or let them go as they are presented to him?

Answer by Dr. C. S. Caverly, Rutland, Vermont.

The health officer is obliged to get the certificate of death and it is his duty to see that it is as complete as possible before he accepts it. He is authorized distinctly not to accept any vague cause as the cause of death.

Discussion by Mr. Burpee.

When a physician has not been called until the decease, can we get a correct cause of death then?

Answer by Dr. H. D. Holton.

To get the cause of death under such circumstances, the health officer can order a post mortem and should always do so.

Discussion by Mr. Burpee.

The statutes say it belongs to the duties of the first selectman.

Answer by Dr. H. D. Holton.

No. The health officer should do it.

Discussion by Dr. E. E. Ellis, Brookfield, Vermont.

Suppose there were no suspicious circumstances, would you advise the health officer ordering an autopsy?

Answer by Dr. H. D. Holton.

Suppose you had been attending that patient and knew he had organic heart disease. You would expect he was going to die suddenly. You could

make out a certificate to that effect. Suppose a man had never employed a physician. He dropped dead. Some say he had heart disease, some say apoplexy. You must make a post mortem to find out the cause of death.

Discussion by Dr. H. H. Swift, Pittsford, Vermont.

I think we all see cases of sudden death or death from more or less uncertain causes where we would not be justified in ordering a post mortem. Shall the health officer look into the case and give the most probable cause of death or put it down as unknown?

Answer by Dr. H. D. Holton.

When you say "probable cause of death" you would be guessing, would n't you? We don't accept any guess work in the classification of these certificates.

Question by J. P. Newton.

The law says no dead body shall be left five hundred feet from the highway; if they do leave them so, can we order them to bury the body just the same as though they were less than five hundred feet?

Question by Dr. E. R. Clark.

I have a rather interesting experience to relate. This experience occurred December 27, 1904. The man was sixty years of age. He had not been feeling particularly well and when he came to town he had promised his wife he would run in to see me. He called at my office that afternoon and I examined him very carefully. After he left my office, he dropped dead

in my presence. I gave the cause of death, apoplexy. It was just an accident that man had been examined; he had no disease of the heart so far as I could determine.

Discussion by Dr. E. R. Clark, Castleton, Vermont.

In my estimation, the registration of vital statistics is one of the most important matters that the health officers have to do with. I have undertaken to see that every certificate that I make out, is filled out to the very best of my ability. I realize it is the important factor of our position to do these little things in the very best, possible way. Many times I am asked what is the use of all this fussing. I explain the reason why and the necessity of it. They readily see the benefits which the people are to derive in time to come and they help me to their utmost, I am sure.

As to the correct cause of death. I had a certificate returned to me quite recently with cause of death broncho-pneumonia. The death of a child which occurred at the same time was stated as uremia. I did not find out until a week later that the woman died from puerperal convulsions. What value

is a certificate returned in that kind of a way? What I want to know is, is it the duty of the health officer to go about and find out the cause of death?

Closing Discussion by Dr. C. L. Wilbur,

I am very glad to see that so much interest has been taken in the important subject of vital statistics, as shown by the discussions and questions on the papers presented. In regard to Mr. Whittier's statement as to the difficulty of obtaining a complete registration of births, I may say that it is easy to do so if you are determined to secure registration of all births and if the people of the community will support you in this effort. We know that many children die soon after birth. Take your certificates of deaths of infants under one year of age and see if the names so reported were duly recorded as births at the time of their occurence, or as soon thereafter as the law requires. If they have not been, then the law has been violated and prosecutions should follow, of course, after proper warning and notification to all concerned. This will do more toward securing complete registration of births than anything else. Merely voluntary compliance with registration laws has always proved ineffective.

I fully agree with what Mr. Stillson has said in regard to the desirability of using the simplest English terms for stating causes of death upon certificates. They should be precise and only a single term should be employed to designate the same disease. At the recent meeting of the American Medical Association, held at Atlantic City, a committee was authorized on nomenclature, and it may be hoped that that committee may recommend a standard nomenclature of diseases which will serve the same purpose in this country as that which the nomenclature of the Royal College of Physicians of London performs in Great Britain.

As to joint causes of death, how shall certain certificates be classified when two or more causes of death are given? We have no universally accepted rules for the classification of several causes. It is a matter requiring professional decision. A town clerk should not assume the prerogative of deciding as to which term should be copied or omitted from his report to the central office, but should copy every word on the certificate forming a part of the statement of cause of death made by the physician.

Reference was made to the terms "chief," and "contributory," and "direct," "immediate" and "remote" suggested to take their places. I would have no objection to any desirable changes on the forms of certificates of deaths, provided all physicians and registrars would understand them in the same sense. We do not use the word "chief" in the standard certificate of death in its exact form as recommended by the Bureau of the Census. The statement is, "The CAUSE OF DEATH was as follows." Following this is a place for the "Contributory" cause of death. In Indiana some changes have been made. Their blank at present gives first the immediate and then the contributory cause. In many cases physicians may report

that the immediate cause of death was heart failure; the remote or contributory cause would be the disease leading up to that condition.*

The completeness and promptness of registration depend primarily upon the interest and efficiency of the local registrar. A prompt registrar and one who sees that the requirements of the law as to prompt returns from physicians are compiled with, will secure complete returns. Delay inevitably means loss of accuracy and missing records. The local registrar should examine certificates as soon as filed with him, and see that missing items are completed if possible. I thoroughly believe that no fee should be given for imperfect certificates, nor should the local registrar receive a fee for a delayed certificate or one not filled out with all the items required by law, providing that it is possible to obtain them.

In concluding I desire to express my appreciation of the invitation extended by the State Board of Health to address this meeting, and of the courteous attention of the health officers present. It is seven years since I had the pleasure of addressing a similar meeting in this city, and I am glad to recognize in the increased and enthusiastic attendance, the manner in which the Annual Schools of Instruction have commended themselves to the people of Vermont as a most important and necessary part of the sanitary administration.

*Since the meeting at Burlington, I have made a special study of the different forms employed upon certificates of deaths in this country and abroad for the statement of cause of death, which I trust may be submitted to the consideration of the registration officials at an early date through the medium of a census pamphlet.

PURIFICATION OF WATER.

GEORGE W. FULLER, C. E., NEW YORK CITY.

I am very glad indeed to be here to-day to speak before the School of Health Officers of Vermont. This school has done and is doing a great deal of practical good in various branches of municipal sanitation. It not only sets forth the theory of the subject, but, what is more important, it provides for the acquirement of practical knowledge on the subject through familiarity with actual applications to various problems which have been worked out elsewhere. The founders of the school should be congratulated upon the success which they have achieved.

I speak these words of appreciation of your work, because I have knowledge of what has been done. When I came to look over the various articles in your Quarterly Bulletin on the subject which I am to speak upon to-day, I found an excellent account of many of the more important filter plants now in successful operation in this country.

Realizing that you have already been well informed as to the main features of this subject, I shall speak to-day upon the subject of water purification, particularly with reference to the new improvements which have recently come up for consideration in this field. I will compare the relative merits of sand and mechanical filters and then I will show you some lantern slides illustrative of a few of the more essential features of recent plants.

Water purification deals with that branch of municipal sanitation by which water supplies are made thoroughly satisfactory, regardless of their source. It deals with ways and means by which the sewage polluted water is freed from disease germs; the muddy water made clear; swampy water made colorless; in other words, the procedures by which a water is made of satisfactory appearance and wholesome quality, so that the public has confidence in it and may drink it with perfect safety.

This subject is a somewhat broader one than that of water filtration. It now includes for some problems, water clarification; in other cases, water decolorization; and in some instances, water softening. Usually there is a combination of features to be dealt with. Rapid progress has been recently made in meeting these varying demands. It can now be stated that purification can now be designed to meet the requirement of practically any problem other than the removal of excessive quantities of salt found in the gas and oil fields.

A few years ago we heard much about water filtration at Lawrence, Mass., where as Mr. Clark has told you, the Massachusetts State Board of Health did much to put this whole subject on a firm footing by conducting an extensive series of tests. The outcome was the establishment of the Lawrence city filter which caused a striking reduction in the amount of