Since R. Pfeiffer and Kruse have seen the influenza bacilli thrive for months in the sputum of tuberculous subjects who were affected by influenza, Baumler thinks it possible that patients afflicted with chronic diseases of the chest, who often travel much, can very easily disseminate influenza. We cannot attribute any epidemiologic significance to this assumption.

After the great pandemic influenza became an endemic disease in innumerable areas of the earth, and especially in the whole of Europe and of North America, and then showed the same fluctuating epidemiologic character, the same puzzling behavior, so far as the time and place of the epidemics were concerned, as other endemic contagious diseases, like scarlet fever, measles, diphtheria, epidemic cerebrospinal meningitis, and others.

This conception of influenza is in harmony with the fact, which has been almost unnoticed up to the present time, that the true endemic influenza, in contradistinction to the pandemic one, is markedly seasonal. Not only the spring epidemic of 1891 and the winter epidemic of 1891-1892, but also the more important epidemic outbreaks of the following years, up to the present day, generally appear in the fall, winter, and spring months, while summer represents the dormant period. We would here call attention to the fact that influenza shows the influence of season (compare p. 533) at the elective point of origin of the great pandemics (central Asia), where, in the fullest sense of the term, the disease is endemic. Almost all the pandemics of this and the previous centuries have started from Russia in the fall, winter, and spring months. Such was the case in the pandemics of 1729, 1782, 1788, 1789, 1799, 1805, 1830, 1833, 1836, and 1889.

We can now recognize also the reasons why the older authors, from Most to Zulzer, were so at variance regarding the influence of season, the laws which govern the areas involved, and the time consumed in the dissemination of influenza. Naturally those younger authors also, who, immediately after the latest pandemic, tried to apply their experiences to influenza as an entity, were led into the same contradictory statement.

We know that influenza presents at least two phases, one pandemic and the other endemic, and they follow different epidemiologic rules.

Even if, as we have shown above, the origin of the later epidemics depended chiefly upon season or location (that is, places where germs capable of further development have remained), this fact, nevertheless, does not in the least affect the "contagious" theory. All the cases of influenza which followed the primary pandemic of 1889-1890, both the continued sporadic cases as well as the occasional local epidemics, were derived from the germs that had been disseminated over the whole earth by the pandemic, and which always and everywhere have multiplied in an exclusively endogenous (endanthropic) manner, and have been distributed by contagion. But we can no more answer the question why, after the cessation of the pandemic, the germs which had everywhere been left behind and had continued to be propagated for months by sporadic cases with lessened virulence (a relative latent period), should suddenly produce marked local epidemics of increased virulence and contagiousness, than we can say why scarlet fever, measles, diphtheria, and epidemic cerebrospinal meningitis should show a similar changeable epidemiologic character. Parsons' statement, that "A periodic multiplication and an increasing virulence, with a converse diminution in these phenomena, is an established biologic law in the life of pathogenic micro organisms," is only a paraphrase and not an explanation.

In view of the fact that the intensity and frequency of influenza epidemics have universally diminished from year to year since the last great trailing epidemic in the winter of 1891-1892, and that the recent winters have gone by without any outbreaks worth mentioning, we may hope that the present septennial influenza period is nearing its termination. The germs that were distributed by the pandemic of 1889-1890 will gradually die, and the influenza vera disappear. It is to be hoped that many decads will pass before a new powerful pandemic will distribute the germs over the earth and render them virulent for years.

As we have considered in the preceding sections the epidemiology of influenza on a large scale, we will now view it in detail-that is, in families, inclosed institutions, in high altitudes, and upon the seas. Following this we shall discuss the important etiologic and epidemiologic factors, such as immunity, period of incubation, morbidity and mortality, the influence of age, meteorologic conditions, etc.