the cases reported in the usual way to the board of health and have increased the accuracy of the information available regarding the prevalence of disease. The collaborating epidemiologist has also distributed to registrars blanks on which they report weekly the names and addresses of the mothers of babies whose births have been registered. Upon receipt of these reports, supplement No. 10 to the Public Health Reports, "The care of the baby," is sent to each mother.

The collaborating epidemiologists in Minnesota, Massachusetts, Maryland, and Kansas keep the bureau informed currently by telegraph of the occurrence of outbreaks of communicable diseases in their respective jurisdictions. In the other States current information of the occurrence of outbreaks of these diseases is not as yet available.

PREVALENCE OF DISEASE IN THE UNITED STATES.

Efficient work on the part of a health department depends primarily upon a knowledge of the prevalence and location of cases of those diseases which it is the aim of the health department to control. This information of disease occurrence is essential to the work not only of the local health departments, but also of State health departments and the Federal health service. No health department, State or local, can effectively prevent or control disease without knowledge of when, where, and under what conditions cases are occurring.

Pursuant to the specific responsibilities imposed by Congress in the act approved February 15, 1893, every effort has been made to keep currently informed of the prevalence and geographic distribution of the preventable diseases throughout the United States. Not only has advantage been taken of existing sources of information and the records available in health departments of States, cities, counties, and townships, but, in accordance with the established policy of the department, an attempt has been made to increase the available information through the education of health officers and practicing physicians as to its importance. Health departments are rapidly coming to realize that it is necessary that they have exact and current information of the prevalence of diseases if their efforts at control are to be at all effective. As a result the extent of the territory in which records of the current prevalence of disease are available is rapidly increasing. It is becoming generally recognized as a truth that efficient health administration and accurate information of the prevalence of disease go hand in hand, and are inseparable.

The available data relating to the prevalence of disease has been currently published in the Public Health Reports for the information of health officials throughout the country.

During the latter part of the year poliomyelitis, or infantile paralysis, became epidemic in New York City and adjacent territory. Because of the great dread in which the disease is generally held and also that all possible measures might be taken to prevent its spread, it was especially important that the occurrence of cases should be promptly known. Officers of the Public Health Service and health officers throughout the country have kept the bureau in

The information

formed by telegraph of the occurrence of cases. thus obtained made it possible to watch the gradual extension of the disease and gave the confidence and assurance that accompanies definite information.

ANTHRAX.

A special inquiry was made to ascertain the recorded prevalence of anthrax in the United States. A letter was sent to each State health officer asking for a statement of the number of cases recorded in his State and the number of deaths registered as due to anthrax. The results are shown in the accompanying table, only those States being included which reported either recorded cases or deaths.

Additional information regarding the source of infection was given as follows:

California. During 1915 one case was said to have been contracted in handling hides shipped from China; one case in handling the carcass of an animal; one case by scratching a toe while swimming in an irrigation ditch; one was an infection in the ankle, but it was not known how it was contracted. The three cases in 1916 were all contracted in the skinning of animals.

Indiana.-Anthrax is not a notifiable disease in Indiana.

Kansas. During 1915 one case of anthrax was reported; in 1916 to September 1, three cases. Of the 1916 cases the source of infection of one was given as due to handling hides. Probable source of infection in the other two cases was not known.

Louisiana. Of the four cases in 1915, one was said to have been contracted from an infected cow; two in skinning carcasses of animals; one occurred in an employee in a blacksmith shop.

Maryland. The case reported in 1915 was said to have been contracted in skinning the carcass of a cow.

Minnesota. The one case reported in 1915 was said to have been of doubtful diagnosis. No post-mortem examination was made. The condition resulted from what was supposed to have been the bite of a flea. There was no previous history of anthrax in the neighborhood.

Utah. Of the two cases reported in 1915, one was said to have been caused by the bite of a flea, the other to have been caused by handling infected animals.

DENGUE.

An outbreak of dengue was reported in Laredo, Tex., November, 1915.

In Hawaii 21 cases were reported during 1915. In Porto Rico there were 13 reported cases, 4 in September, 5 in October, 1 in November, and 3 in December.

DIPHTHERIA.

Diphtheria is present throughout the country. Cases are found practically in every community of any size at some time during the year. It attacks children more frequently than adults. Usually, therefore, more cases are likely to occur in populations having many children, and, conversely, few cases may naturally be expected in communities with few children.

Before the days of the bacteriological diagnosis of the disease only those cases were recognized as diphtheria which were characterized by a well-developed membrane in the throat or nose. These included the more severe cases. With bacteriological diagnosis made easier, especially in the large communities, many cases which formerly would not have been recognized are now known to be diphtheria. These include many of the clinically milder cases, without typical membranes or with no membrane at all. With the aid of bacteriological diagnosis a larger percentage of the cases of diphtheria is being recognized and recorded, but there are undoubtedly many mild cases which are still unrecognized as being diphtheria. This recognition of the mild as well as the severe cases should have the effect of giving a lower indicated fatality rate, and this has been the effect wherever satisfactory records of the prevalence of the disease have been secured.

In the comparison of records of diphtheria in different communities difficulty is met because of the varying completeness with which cases of the disease are reported. The differences found in making comparison are as likely to be due to differences in the completeness of the records as they are to differences in the prevalence of the disease. This condition would be changed under uniform requirements for notification of diease and uniform practice as to the enforcement of these requirements.

Of the States having health departments which attempted to secure information of the prevalence of disease within their jurisdictions for the calendar year 1915, the highest reported case rate for diphtheria was that for New Jersey, with a rate of 2.41 per 1,000 inhabitants. The next highest rates were in Virginia and New York, with rates of 2.39 and 2.06, respectively. The lowest reported rates were in Nevada, Louisiana, and Wyoming, with 0.01, 0.08, and 0.11, respectively.

Of the cities of over 100,000 population, the highest reported case rates were in New Orleans, St. Louis, and Boston, with 5, 4.84, and 3.91, respectively, per 1,000 population. The lowest reported case rates were in Seattle, Louisville, and Denver, with 0.21, 0.72, and 0.91, respectively.

For the year 1915 the highest indicated fatality rates were in Nevada, Alabama, and Arizona, with rates of 100, 34.58, and 31.19,

respectively, per 100 reported cases. The fact that Nevada reported a very low case rate and a 100 per cent fatality rate indicates that reporting of cases was very incomplete. The lowest fatality rates reported were in the District of Columbia, South Carolina, and West Virginia, with 4.35, 4.67, and 4.94, respectively, per 100 cases reported.

The highest fatality rates recorded in the larger cities were in Fall River, New Bedford, and Albany, with indicated rates of 20, 17.33, and 14.29, respectively, per 100 cases reported. The lowest rates were in Richmond, Denver, and Seattle, with 1.92, 2.62, and 2.86, respectively.