There is a scientific paper "Cancer Trends During the 20th Century" published in the Journal of the Australasian College of Nutritional & Environmental Medicine, Vol. 21, No. 1, April 2002, pages 3-8. It shows interesting graphs for cancer deaths (mostly in Sweden which had good quality records) over most decades of 20th century:

<img src="http://www.acnem.org/images/journal/21-1_april_2002/21-1_halberg-fig_3.gif">
Figure 3. Skin melanoma mortality in Sweden since 1912.

<img src="http://www.acnem.org/images/journal/21-1_april_2002/21-1_halberg-fig_4.gif">
Figure 4. Lung cancer death rates in Sweden.

It shows clearly simultaneous rapid rise in the rates for both types of cancers in the fifties, just as the atmospheric nuclear tests (especially in neighboring USSR) were reaching their peak (these tests were banned not long after the cancer rates rise started). The skin and lungs would be particularly sensitive to airborne radioactive particles, hence the rates of these two cancers show high degree of temporal correlation. Note also how there is another little upward jolt in the rates, especially striking for melanoma rates, in late 1980s, right after the Chernobyl nuclear plant meltdown. In the article they also show how this same correlation (between melanoma & lung cancer) manifests in spatial dimension as well, when data is compared by counties in Sweden, which is to be expected since the air current patterns would deposit radioactive fallout differently at different locations.

Another interesting bit buried in the paper is how the statistics around this critical period were difficult to get from the government "health" bureaucrats and how the more recent official way of counting the cancer death rates, the so called 'age adjusted mortality', is artificially making the increase appear smaller (citing discussion after Fig 13):

" First of all, data before 1955 has always been locked out from publicly available databases. Secondly, and even worse, trends for obviously exposure-time-dependent cancers are effectively neutralised by use of age-standardised ratios (ASR). This procedure assumes that the increasing cancer incidence is a natural effect of growing old and thus the age standardised mortality will stay the same although the population gets older. By doing this, the responsible institutions can show to the authorities that the mortality is in control and in effect not increasing at all despite the fact that it is. "

This same statistical gimmick with 'age adjusted rates' was also used by CDC to hide the fact that lung cancers in USA have been on the rise for decades, being now at their maximum, even though the smoking has been declining also for decades, being at its minimum now. The sneaky little mathematical trick of scaling the rates for different years using different multipliers, made the actual increase appear as a small decrease in the "new math" (since they multiplied the rates of more recent years with smaller factors), which CDC "scientists" then attributed to the decline in smoking rates.

Further the asthma rates rose precisely as the smoking declined (and vaccinations increased):

<img src="http://www.acnem.org/images/journal/21-1_april_2002/21-1_halberg-fig_7.gif">
Figure 7. Asthma prevalence among Swedish 18-year-old males and the rejection rate at military conscription test
due to asthma in Finland.

<img src="http://www.acnem.org/images/journal/21-1_april_2002/21-1_halberg-fig_8.gif">
Figure 8. General asthma prevalence in the Swedish population.

Another interesting graph is of the statistics for the locations of melanomas & other so called sun-related skin cancers, showing clearly that they occur mostly in locations the least exposed to the sun, while locations the most exposed to sun (head, arms, legs) have the fewest cancers. In other words, sun is protective against skin cancers (induced by radioactive fallouts & possibly EMF).
<img src="http://www.acnem.org/images/journal/21-1_april_2002/21-1_halberg-fig_15.gif">
Fig. 15. The combined distribution of dysplastic naevi and melanoma (‘dots’) over the human body.