The subject is so vast and involved that it cannot be discussed with any approach to completeness in a single essay. Suppose, however, that we glance at the theory which regards cancer as due to a microbe which in some mysterious ways gains admission into the human body, lying for a time dormant, but liable under appropriate stimulation to be awakened into malignant activity. We know at the outset that if any such germ of disease exists, it has thus far escaped visual recognition. No human eye can be said with certainty to have seen it, even when aided by the most powerful microscope; but this may be due to the fact that, like the germ of certain other diseases, it is so minute that it lies beyond the range of human vision. There are, however, certain facts pertaining to the disease which have significance. We have already seen that in a given country there is a kind of uniformity in the number of those dying from the disease from year to year; but another phenomenon relates to the unequal pressure in difference countries of the causes of the disease. 1. THE DEATH-RATE FROM CANCER APPEARS GREATLY TO VARY ACCORDING TO RACE AND ENVIRONMENT. CANCER DEATH-RATE IN DIFFERENT COUNTRIES PER 100,000 POPULATION --------------------------------------------------------------- | Five-Year Periods. | Switzerland. | England. | Italy. | |--------------------------|--------------|----------|--------| | 1886-1890 .. .. | 114 | 63 | 43 | | 1891-1895 .. .. | 122 | 71 | 44 | | 1896-1900 .. .. | 127 | 80 | 51 | | 1901-1905 .. .. | 128 | 87 | 55 | --------------------------------------------------------------- Here is the record of a period of twenty years. These differences of proclivity to cancer are exceedingly curious. Can the reader perceive why they exist? The rate in England is quite 50 per cent. higher than that of Italy. If we explain this by the hypothesis of greater skill in detecting the disease, what are we to say of the cancer-rate in Switzerland, which is 50 per cent. higher than that of England? But here is another curious fact. The United States census of 1900 permits a contrast of the mortality of cancer according to the birthplaces of mothers of those attacked. Here, for instance, is the death-rate from cancer and tumour of persons of different nationality, calculated in three sections of the country--the rural districts of the registration area, the cities of the same section, and the cities outside the registration area. DEATH-RATES IN THE UNITED STATES FROM CANCER AND TUMOUR PER 100,000 WHITE POPULATION, ACCORDING TO THE BIRTHPLACE OF THE MOTHERS OF PATIENTS