Occupation also plays a still but little under stood role; thus in England from 1890 to 1892, the death rate per 100,000 from cancer was 67 in clergymen and 265 in chimney sweeps. The irritation resulting from the soot may explain this difference.
Increase of Cancer.—There is a widespread opinion that cancer is increasing rapidly in frequency of occurrence. This question has been subjected to elaborate statistical analysis by Hoffman in a volume on The Mortality of Cancer Throughout the World.) Hoffman comes to the conclusion that there is a gradual increase in the number of cases of cancer oc curring in most of the countries of the world, but that this increase tends to reach a certain maximum and not to pass that point. This maximum has already been reached in Switzer land, where the cancer death rate is now 124 per 100,000, while in the United States it is only 74 per 100,000. There are, however, many authorities who consider that this increase in cancer is due not to an absolute increase in the number of persons with cancer, but to an in crease in reported cases; in other words, that improvements in diagnosis and in our under standing of the nature of the disease and the increased accuracy of statistics have resulted in an increase in the number of cases of cancer which are reported. The matter cannot be con sidered as finally settled, however; and only greatly improved statistical reports from dif ferent countries can throw any light upon the problem.
Heredity.— The question of heredity in cancer has long occupied the attention both of statisticians and of workers with experimental cancers in animals. While there is no doubt that an individual family may show a large number of cases of cancer for one or two i generations, this is by no means a proof that the disease itself is hereditary. It is not even a proof that a liability to the disease is heredi tary. If we realize that out of every 100,000 people living in the United States, 80 will die from cancer, it is easy to see that, as age increases and the frequency of cancer becomes greater (so that between the ages of 45 and 65 one woman in 6 and one man in 12 will die from the disease) the occurrence of several cases in a family, especially if it be long lived, is not astonishing. It is evident that if cancer occur so frequently, some families are bound to have a larger number of cases than others, merely by chance. This point is illustrated in
the °cancer villages" which will be discussed in the next paragraph. The problem of heredity in cancer has interested the life insurance com panies, also, because if cancer be hereditary, it is very important that no one in whose family cancer has occurred shall be accepted by the companies. But a careful analysis of a large number of persons insured shows that the lia bility of death from cancer is no greater be cause that disease has occurred in one's family than it is if no cancer death has been recorded; so that the insurance companies do not regard a history of cancer in the ancestry as any evi dence that cancer will occur in the descendants.
It has been shown, however, by breeding mice of cancerous ancestry, that is, by selecting both father and mother from a highly can cerous strain, that the amount of cancer occur ring in the descendants after many generations of such selected breeding, is apt to be about twice as much as occurs in the stock in which no special selection was made. Such a con centration of cancer heredity in man is quite impossible, as the stock would die out long before the 8 or 10 cancerous generations could be produced.
Contagion in Cancer.— It is a popular su perstition that cancer is contagious, and that it is very dangerous to nurse a person with an ulcerating cancer; but there is not the slightest foundation for this belief. There is no recorded evidence of a case in which cancer has been transferred from one person to another or in which a surgeon has contracted the disease dur ing an operation on a cancerous patient. If cancer were a germ disease, it would probably be very easily transmitted in this way; but there is no evidence that a germ has anything to do with the development of the tumor; and the only way in which the disease can be trans ferred is by direct grafting of a piece of the growth from a cancer patient to a healthy person. It is very difficult even to graft a cancer from one mouse to another. Often hundreds of animals are inoculated in order that one or two cancers may be grown. Ani mals bearing cancer and healthy animals have been kept together in large numbers for long periods, and not the slightest evidence of any transfer has been shown.