marked attack is again somewhat more dangerous. The female sex throughout life is more liable to scarlet fever than the male sex, but the attacks in males, though fewer, are more likely to terminate fatally.

The proportion of fatal cases to attacks of scarlet fever cannot be accurately stated, owing to the large number of unrecognised cases of very mild type, often without skin eruption and with very little desquamation. If all such cases of infectious sore-throat without eruption, which are by no means uncommon in adults or those who have been partially protected by a previous attack, could be included, the case mortality (proportion of deaths to attacks) would probably be found to be not greater than I or 2 per cent. These very mild and unrecognised cases are, doubtless, most frequent sources of dissemination of infection, and the fact of their being true scarlet fever cannot be doubted.

For the 10 years 1871-80, the death-rate from scarlet fever per 1000 living at all ages was about 0.7. Under 5 years the death-rate was 3.5 per 1000; between 5 and 10 years it was 15 per 1000, decreasing in the next quinquennium to 0.32 per 1000. During the eight years 1881-88 the average death rate in England from scarlet fever was 0.36 per 1000 living at all ages.

Unlike small-pox in unprotected communities, scarlet fever is a disease from which very many people altogether escape. The importance of saving young children from attacks of scarlet fever has been well expressed by Dr. Whitelegge."

"In shielding a child against infection during the first few years of life there is a double gain; every year of escape from scarlet fever renders him less and less sus

Age, Sex, and Season in relation to Scarlet Fever. Transactions of the Epidemiological Society, vol. vii.

ceptible, until finally he becomes almost insusceptible; and, secondly, even if he should ultimately take the disease, every year that the attack is deferred reduces the danger to life which it brings. In other words, attacks of scarlet fever become both less severe and less frequent with every year of age after the fifth. Up to the fifth year the liability is less (than in the fifth year), but the risk to life in case of attack is very great." The same reasoning applies with almost equal force to measles, whooping cough, and the other infectious complaints of childhood.

Overcrowding and insanitary conditions in houses tend to aggravate the severity of scarlet fever attacks, and possibly aid in their dissemination, but can have no influence per se in originating an outbreak.

Scarlet fever is most prevalent and most fatal in the autumn, in the months of October and November. Two curves may be formed; one expressing the weekly or monthly deaths as percentages of the average weekly or monthly mortality throughout the year (fig. 69); the other expressing the number of weekly or monthly cases as percentages of the weekly or monthly average of cases throughout the year (fig. 71). These curves correspond very closely, but Dr. Whitelegge has noted that the mortality-curve rises less and falls less above and below the mean than the case-curve-which would imply that when most prevalent scarlet fever is least fatal and vice versa. There is, at least, a strong probability in favour of this view, as the number of mild cases is usually greatest when scarlet fever is most prevalent.

FIG. 71.-Seasonal curve of cases of scarlet fever stated as percentages of monthly average, based upon 23,000 cases notified in nine large English towns, 1885-6-7. (B. A. Whitelegge, Epidem. Soc. Trans., vol. vii.).

Measles.

A specific infectious fever with an incubation period of 9 to 12 days. The contagion is given off from the breath and skin of the patient during the whole period of illness (3 or 4 weeks), and the catarrhal stage preceding eruption is especially infectious. It is for this reason that epidemics of measles are so difficult to control. The infection is not widely diffusible in the air, but clings to clothes and garments with a persistency little if any less than that of scarlet fever.

Measles is a disease of infancy and early childhood, and is very fatal to young children, chiefly owing to the frequency of pulmonary complications and sequelæ. Adults unprotected by a previous attack are also susceptible, but the disease is so universal in this country that few children escape from it. The mortality from measles is greatest under 3 years of age; after 5 years of age the mortality is enormously diminished. During the 10 years 1871-80, the death-rate from measles averaged about 0.38 per 1000 living at all ages; under 5 years of age the death-rate was 2.5 per 1000, and between 5 and 10 years only 0.2 per 1000. In the 8 years 1881-88, the average death-rate in England from measles was 0.42 per 1000 living at all ages. Both sexes are equally

liable to attack, and the case-mortality is about the same for both. In this disease the case-mortality, or proportion of deaths to attacks, is greatly affected by overcrowding and insanitary conditions generally. In the overcrowded houses of the poor, amongst badly nurtured children, the proportion of deaths to attacks may be as much as 20 or 30 per cent., and is, no doubt, intensified by the neglect of the parents to provide suit

able warmth and nourishment for the sufferers from a disease which they think of little moment. In healthy houses, well-nourished children almost invariably make a good recovery.

Measles is most prevalent and most fatal in the winter months of November, December, and January; but it also tends to become somewhat intensified in the late spring (May and June), see p. 384, fig. 69.

Measles epidemics tend to recur in large towns about every 3 or 4 years, with the fresh appearance of susceptible children.

Rötheln, Rubella, or German Measles.-This is a specific infectious fever, propagated by a specific contagium, and not a hybrid between measles and scarlet fever, from either of which diseases it is entirely non-protective. It has an incubation period of 10 to 14 days, and is infective during the whole course of illness (7 to 14 days). It is not a disease of common occurrence, and the illness produced is almost invariably very mild. It is probable that children and young adults are most susceptible.

Whooping Cough.

This is a specific infectious disease, the infection being given off in the breath and secretions from the lungs. It is probably not carried far in the air, but clings pertinaciously to articles of clothing. The period of incubation may last from 1 to 3 weeks, and the period of infectiveness is usually not less than 6 weeks from the onset of cough, and may be longer.

Infants and young children are specially susceptible, and few escape attack. The younger the child the