TABLE 2.-Weekly deaths from gastrointestinal diseases in Berlin during the summer of 1905 (after Liefmann and Lindemann).

We see from the foregoing that the march of infant mortality in the first part of the summer is characterized by the great number of deaths with acute symptoms and short duration (24 to 48 hours), while later the high, broad, and slowly descending curve is largely produced by the deaths of infants who succumb to more or less subacute affections, mainly of the gastro intestinal tract.

The influence of heat upon the mortality of infants in the first part of the summer is apparently immediate. In the latter part of the summer this no longer seems to be the case. At first sight one would be inclined to state that, while some causal relation is manifest, the late summer mortality of infants is not directly influenced by the heat, which, therefore, can stand in no immediate relation to this phenomenon.

A closer examination, however, shows that we can, by no means, suppose that infants are necessarily relieved from the action of heat by reason of comparatively long remissions of the temperature in the latter part of the summer. If we are to examine the effects of heat upon infants, we can not arrive at conclusions by considering outdoor temperatures only. The nursling passes most of his time indoors, and it is the indoor meteorological conditions to which he is mainly subjected.

Indoor Temperatures.

The credit of calling attention to the importance of indoor temperatures belongs to Flügge (11), who, in 1879, published a series of observations upon indoor temperatures in midsummer. He found that the temperature indoors was dependent upon the degree of heat received by the walls from the sun.

The diffusion of the heat through the walls was found to be relatively slow, so that their inner surfaces reached their maximum temperatures in the following order: East wall, 9 p m.; south wall, 12 m.; west wall, 3 a. m.

The effect of the radiation of heat from the inner surfaces of walls, protracted far into the night, was to maintain the indoor temperature at a level considerably in excess of that of the external air. During a relatively cool summer Flügge found that the mean indoor temperature might exceed the exterior by as much as 9.5° C. (17.1° F.).

Other observations on the indoor temperature have been made by Meinert (41, 42), Hammerl (16), and Rietschel (55), in Germany; in this country by Chapin (4), Knox (29), and Helmholz, (19).

Meinert (41, 42) observed an average excess of 8.8° C. (16° F.) of the indoor over the outdoor temperature in a room in Dresden in which an infant had died of summer diarrhea.

Hammerl (16) made observations in the dwellings of workmen and found that the average mean indoor temperatures were greatly in excess of the outdoor. In one instance the temperature never fell below 24° C. (75.2° F.) for a period of over 14 days and often rose to 32° C. (89.6° F.), and, on one occasion, to 35° C. (95° F.) with outdoor temperatures that never exceeded a maximum of 23.7° C. (74.6° F.).

In the cool summer of 1910 Rietschel (55) made observations. with maximum and minimum thermometers and recording thermographs in dwellings in Dresden, in which, for the past 5 years, a high infant mortality had been recorded.

He states:

I admit frankly that the results astonished me. I had indeed supposed that in such dwellings temperatures of considerable height might prevail in summer, but I had never expected that such excessive temperatures could be maintained, not only daily but weekly, with an outdoor temperature relatively so low.

In some instances the temperature rose to 35° C. (95° F.) and 37° C. (98.6° F.), with outdoor maximums of but 23.5° C. and 21° C. (74.3° F. and 69.8° F.). During the hot summer of 1911 he reports an instance in the home of a workman of maximum temperatures of 38° C. (100.4° F.) with a minimum of 30°C. (86° F.).

Similar measurements have been made in this country by Chapin (4), Knox (29), and Helmholz (19) during the summer of 1912 with like results. Helmholz observed, in the stockyards district of Chicago, maximum indoor temperatures 30° F. in excess of the outdoor maximum. Minimum temperatures below 80° F. were rather infrequent, occurring only 83 times out of 1,374 maximum and minimum readings.

It is clear, from the foregoing, that infants are often obliged to endure, for considerable periods, temperatures which are greatly in excess of those of the external air. Moreover, the effect of these temperatures is often enhanced by excessive clothing and rubber

diapers. It is a too frequent experience to find, in crowded tenement homes, the busy mother keeping the baby in the kitchen, often near the stove, in order to have it close at hand.

Effects of Humidity.

The influence of a high degree of moisture in the air in increasing the effects of heat is well known. It would naturally be expected that hot days with high humidity would show the greatest increase in infant deaths. The observations on this point are inconclusive. Days on which great infant mortality has been recorded have, for the most part, been hot and dry.

Meinert (41), Prausnitz, Liefmann and Lindemann (37), and others point out that the relative humidity of the external air is by no means an index of the moisture to which infants are exposed. In the dwellings of the poor, where the great majority of infant deaths. take place, the infant passes a large part of his time in narrow and usually crowded quarters, where the relative humidity may be greatly raised by the lack of ventilation, the moisture derived from the breath and skin of numerous inmates (frequently 5 or 6 in a single room), and the water evaporated in cooking, ironing, and washing.

Indeed, Kubly's (31) observation (cited by Meinert) showed that in the dwellings of the poor in Dorpat a mean relative humidity of 83 per cent was found in 223 observations and in 30 instances 90 to 98 per cent was recorded. Knox's (29) observations in Baltimore during the summer of 1912 showed the relative humidity in such dwellings to vary between 70 and 80 per cent.

Circulation of Air.

It is evident that circulation of the air is of great assistance in eliminating heat from the body by accelerating evaporation. I will cite only the recent experiments of Flügge and his coworkers, Paul, Ercklentz, and Heymann (13). They showed that in stagnant air symptoms of heat retention begin to make their appearance at temperatures of 24 to 25° C. (75.2° F. to 77° F.), which, however, disappeared when the air was set into motion by a fan.

If, then, heat is to be regarded as a direct factor in the causation of summer infant mortality, we would expect to find an increased number of deaths of infants in houses and in sections of cities where conditions are unfavorable to the circulation of air. Such, we find, is the case.

Ballard (1) in his extensive investigations of summer mortality in England noted that when the wind had free access to houses the summer infant mortality was low, while it was high in those cities, or sections of cities, where, from the disposition of the streets or houses, they could not be swept by the prevailing winds in summer.

Meinert (41), to whom belongs the credit of being the first to investigate in Dresden the relation of housing conditions to the summer mortality of infants, found that in practically every instance in the case of 580 infants dying in the 11 summer weeks of 1886. the conditions were such as to prevent the free circulation of air in the rooms in which these children lived. In only one instance did any of these infants die in a house exposed freely on all sides to the wind. The infants died in those dwellings which, from their low-lying situation, their location in the rear of other dwellings, the presence of courts, narrow streets, and shut-in style of architecture, were denied the access of cooling breezes or, in which from the disposition of the windows, through ventilation was impossible.

Thus on low-lying Hechtstrasse, with shut-in houses and many courts, 18.49 per cent of the living children under 1 year died; on high-lying Kiefernstrasse, with a more open style of architecture, only 2 per cent; upon Johan Meyer Strasse, with model workmen's homes, open to the air on all sides, none.

Prausnitz (52) points out that in the absence of through ventilation the only exchanges which can take place between the inside and the outside air arise through differences in their temperatures.

In the absence of through ventilation these differences of temperature are so slight, in hot weather, that stagnation readily takes place. He and his coworkers investigated in Graz the location of windows in dwellings in which infants had died of summer diarrhea during the years 1903 and 1904. A dwelling was regarded as susceptible of through ventilation when it possessed windows in opposite walls, as partially so when the windows were in walls at right angles to each other, and as not susceptible when windows in one wall only were found.

They obtained the following results:

It will be seen from the above that approximately 80 per cent of these dwellings did not present conditions favorable to a free circulation of air.

Other Housing Conditions.

Besides Meinert's former work other careful studies of housing conditions in relation to housing have been made. Among these may be especially mentioned those of Willim (63) in Breslau, of Liefmann (34), and of Kathe (25) in Halle, and of Liefmann and Lindemann in Berlin. These investigations all bear testimony to the great

influence housing conditions have upon infant mortality, and particularly the conditions favoring high indoor temperatures or the absence of free circulation of air.

While it is true that the conditions were often such as would also favor the dissemination of infectious disease, we encounter the paradox that certain conditions, intrinsically unhygienic, have a distinctly favorable influence on the summer mortality of infants.

Thus it was found by Meinert, Willim, and Liefmann and Lindemann that infants living in basements and cellars show a smaller increase in their summer death rates than do breast-fed infants; while, with respect to the other stories, the death rate usually increases as we go up, being found highest next to the roof. (Meinert found the greatest number of deaths in the ground and first floors.)

Liefmann's and Kathe's investigations in Halle showed that the summer mortality of infants was rife, not only in certain streets, but in certain houses. Thus Liefmann found that out of 380 streets which Halle possessed at that time, no less than 141, or over 37 per cent, had no infant deaths.

We can also readily conceive from the character and situation of the houses in which infants die that, once they are heated through by the warmth of the early summer, the heat stored in the earth. pavements, and wall-, in their shut-in location, prevents any substantial cooling off in the late summer, in spite of periods of moderately cool weather. In the early summer, before there is much stored heat, remissions in the external temperature have a much greater effect.

Effects of Heat in Causing Death.

It must be clear from the foregoing that the greater the summer heat the greater the number of deaths and that it is an indoor and not an outdoor climate to which infants are exposed. There remains for discussion the manner in which their deaths ensue. There are two hypotheses by which these are usually explained:

I. Owing to the fact that the great majority of infants dying in the summer are bottle fed, the deduction is logical that the onus of their deaths must be placed largely upon the food most generally used in the artificial feeding of infants, namely, cows' milk, which becomes so readily spoiled by the summer heat when carelessly produced and handled.

We find two views generally held to explain the pernicious action of cow's milk in the summer:

1. That as a result of careless handling and inadequate cooling on the part of the producer and the consumer, the germs which always contaminate cow's milk undergo such proliferation that the ordinary saprophytic bacteria in milk or their toxic products endanger the life of the artificially fed infant.