The population living in unincorporated towns and villages has for the second successive year attained the lowest death rate of any group and has just managed to maintain the low record of the preceding year. Like the preceding class some deaths. properly chargeable to the group have occurred in other places and after adding these and deducting deaths of residents of incorporated places there is a net increase of seventeen, giving a death rate of 6.2 per 100,000, which at that is the lowest of any group of the series. Several outbreaks have occurred during the year, two of which, Wellsville and Millbrook, have been previously mentioned. Others noted were in the towns of Colonie and Bethlehem, Albany county, Skaneateles, Onondaga county and in several institutions to be noted later. Most of the Colonie cases were found to be due to the use of water from a well used in common by a number of families in the unincorporated village of Menands. The few cases later in the year contracted the disease. through the use of Watervliet water. The Skaneateles cases were located in the unincorporated village of Skaneateles Falls but no definite evidence as to the source was obtained. A milk-borne outbreak of sixteen cases occurred in the town of Bethlehem, Albany county, in November due to an unrecognized case on a dairy farm. Outbreaks occurred in the following institutions. not under state management; Trudeau Sanatorium (August); Hauppage Orphan Asylum, Smithtown; Colored Orphan Asylum, Smithtown; Wayne County Home, Lyons. In the latter two the sources of infection were definitely traced. The colored orphanage cases were found to be due to eating clams taken from the shore near a sewer outlet. The Wayne County Home outbreak of ten cases, four of whom died, was caused by the habit this institution had of drinking its own sewage in a highly diluted form. There fore shortly after a case of typhoid fever came to the home others living there began to come down with the disease. An investigation of the water supply and sewerage systems disclosed the fact that the sewers discharge into the Erie canal about thirty-five feet from the intake of an auxiliary water supply pumped through the same pump as the supply used for drinking purposes. The sources of the Trudeau and Hauppage outbreaks were not definitely ascertained. In each of them one of the early cases had

recently come to the institution but it was impossible to connect either of these in any way with the food supply or to account by contact with the explosive nature of the outbreaks. In the Hauppage institution a chronic carrier was found in addition to the possible early carrier referred to but, it was not possible to trace the route of infection if she were the true source. The report of the Trudeau outbreak is given elsewhere in detail.

As previously mentioned, the great bulk of the cases in this group were sporadic cases or groups of two or three. Two hundred and thirty cases were investigated during the year, in 103 of which the modes of infection were definitely ascertained. The most common modes of infection were through water and by contact with a case or a carrier. Of the forty-four cases ascribed to water, 10 were due to municipal supplies, the remainder to wells and, in a few instances, to drinking directly from polluted streams. An almost equal number, 42, were ascribed to contact either with cases or carriers, mostly the former, perhaps on account of their being more apparent. It is very strikingly evident, however, that too little attention is paid by physicians to the possibility of contact infection. Too frequently the care of the primary case devolves upon another member of the household not acquainted with the principles of medical asepsis and blissfully ignorant of the potential dangers of the position. Immunization is not extensively practiced and too frequently the "nurse" is also the cook, with resulting disaster to other members of the household. This sequence of events, while by no means confined to the rural districts, is relatively more frequent there. It is due probably to the lack of trained nurses, the remoteness of hospitals and to a failure to promptly immunize other members of the household. The former may be overcome to some extent by the appointment of county nurses, the latter by the education not only of physicians, but the public at large, in the efficacy and innocuousness of this measure.

There remains to consider 75 cases which occurred in state institutions during the year not included in any of the areas considered. Eight of these institutions had one or more typhoid cases during 1917, but the bulk of the cases occurred in four of these. In the Gowanda State Hospital cases began appearing late

in 1916 and with increasing frequency during January 1917, after which the number rapidly subsided. The outbreak was investigated with inconclusive results. A small outbreak of seven cases occurred at the Raybrook State Tuberculosis Hospital in October and November. Investigation failed to disclose a carrier in the institution but did discover an unlicensed labor camp on the watershed of the institution's water supply. No case or carrier was discovered among the laborers but as there had been changes in the personnel of the camp the possibility could not be ruled out. Both here and at Trudeau the staff and patients were vaccinated against typhoid though at Raybrook only those patients were included who were in fair physical condition. At neither institution have there been any apparent ill result from this procedure.

At the St. Lawrence State Hospital a carrier was found in the dining room of the institution who was held responsible for the occurrence of 12 cases in August and September. At the Central Islip State Hospital cases have occurred with considerable frequency the past two years but without any well marked outbreak. The cases have been confined almost entirely to certain wards occupied by demented patients where the habits of the inmates are conducive to contact infection. An outbreak also occurred in May in Manhattan State Hospital, the cases and deaths being included in the New York City figures. Following this outbreak the State Hospital Commission recommended at the suggestion of this Department the vaccination of all inmates under 50 years of age who had not had typhoid and the vaccination of all new admissions. It is hoped that strict adherence to this procedure will much decrease the typhoid incidence in these institutions and lessen the number of the often difficult and sometimes baffling outbreaks that occur in them.

Vaccination should be practiced more extensively in other institutions to prevent serious results consequent upon sanitary neglect or the accidents due to carriers. In this connection it has been somewhat surprising to discover the small number of hospitals that require their nurses to be vaccinated — but 16 out of 48 replying to a letter requesting this information. Three others urge the procedure and have been successful in

inducing a high percentage to accept and four others require vaccination of nurses attending typhoid cases. In many instances the excuse is given that they treat so few typhoid cases that it is regarded as unnecessary to vaccinate. This point of view entirely overlooks the fact that many cases of typhoid are not recognized as such until some time after their admission and that typhoid vaccine is of doubtful efficacy after the individual has been infected. There is not even this excuse, however, for hospitals having a large number of typhoid admissions every year.

The seasonal incidence for 1917 was the same as for almost every previous year of record - a gradual decline in the number of cases until June after which there is a rapid rise until September following which occurs a sharp fall which in the succeeding months becomes more gradual. Though occasionally modified to the extent of peaking in August or October or reaching a minimum in April or May, it is quite remarkable how closely the curve each year follows an average of the whole. This of course is not peculiar to New York State but is practically universal except for individual communities. The favorite explanation of this fall and rise is, in the largest cities, "vacation typhoid " but this fails to explain the synchronous rise in village and rural typhoid and intensive investigations have shown that the relationship is accidental rather than that of cause and effect. Flies, ground water conditions, increased milk infection and other more or less plausible explanations have been offered but it seems that the question is worthy of more adequate study. In 1917 more than 43 per cent of the up-state cases had their onsets in the three months, August, September and October.

In conclusion, it may be stated that while excellent progress has been made in the past, it is becoming more apparent each year that further reductions can only be hoped for through intensive efforts. The great reduction in the typhoid rates since 1900 has been due in greatest part to the activities of the Engineering Division in improving water supplies and installing sewerage systems. It is for them to hold these gains and effect such further improvement as may be possible through the work that lies within their province. The opportunity is not as great as it

was a decade ago — that is, the results cannot be so dramatic but the possibilities are doubtless much larger than many surmise. Because typhoid rates are much lower than prior to improvements in their water supplies many cities are content and ascribe their typhoid to anything but water. While these improvements have frequently served to unmask other causes, yet the lowering in the typhoid rates succeeding further improvements have shown that water supplies considered satisfactory by ordinary laboratory test were really the vectors of a considerable amount of the "residual" typhoid.

However, the reduction in the future as in more recent years will depend most upon the work of the health officer. It is principally a matter of stopping numerous small leaks and strengthening weak points before they break. Among the latter, aside from municipal water supplies protected by chlorination only, the most important is probably the unpasteurized milk supply. While but three up-state outbreaks were definitely traced to milk last year it is extremely probable that more or less infection is transmitted in this way without focusing attention on the milk supply and without assuming the proportions of an outbreak. Neither of these two sources affect in any considerable degree the small village and rural typhoid. As previously mentioned the reports of cases investigated have pointed most clearly to contact with previous cases as the commonest mode of transmission. There have been a few small outbreaks definitely traced to infected water supplies but the bulk of the cases it has been impossible to ascribe to any definite source. The ways by which typhoid may be transmitted are perfectly apparent in practically every small village and on most farms. In the main they are the improperly located, constructed or protected well, the non-fly-proof privy, the leaching cesspool, the untrained nurse and the neighborly habits of the people. The latter results in an unusual opportunity for direct and indirect contact which is often impossible to trace with certainty. These are all matters dependent upon individual actions and are susceptible of correction, in the main, only through persistent educational effort. That a large degree of success has attained such