Now suppose a little bomb whose cloud will remain in the troposphere has been exploded at one of the United States test sites. The Nevada test site is at a latitude of 37°N and the Pacific test site at 12°N. In these middle latitudes, in the troposphere, the winds blow mainly from west to east with an average speed of approximately 20 miles an hour. There will be a slight southerly or northerly motion on top of this. But by and large the radioactive cloud will stay in a pretty narrow band around the latitude at which the explosion took place. After the first few hours, when the close-in fallout has dwindled, the radioactive particles remaining in the cloud are too light and too fine to fall any more under the action of gravity. At this point the weather becomes important. Rain, fog, or mist captures the radioactive particles, and returns them to the ground in the rainfall. This results in the so-called tropospheric fallout.[12] The average time for this fallout to occur is approximately two weeks to a month. During this time, while staying more or less in the latitude of the explosion, the radioactive particles may actually have encircled the earth. The clouds of the big bombs rise high into the stratosphere. The winds in the stratosphere do not blow so predominantly in a latitudinal direction. What is more important, they stay in the stratosphere for years, in which time the radioactivity is distributed to all areas of the globe. The fallout from the big bombs is thus really world-wide. The tropospheric fallout takes about a month. The stratospheric fallout takes 5 to 10 years. The reason for this difference is the weather, or rather the lack of it. In the stratosphere there is no rain or fog to catch the radioactive particles and hence no effective mechanism for producing the fallout. In fact, since the radioactive particles are too fine to fall by gravity, they must simply wait until some turbulent motions impel them downward back into the troposphere. This process requires a long time. That rainfall is the most important mechanism for producing the world-wide fallout has been shown by examining the fallout in certain dry regions of southern California and South America. In every case the fallout was found to be considerably sub-normal. In one place in Chile, where there is never any rain, the fallout was found to be only one per cent of what might be expected on the basis of the average fallout at the same latitude. In regions having at least a few inches of rain per year, the fallout tends to be proportional to the rainfall on the average. However, the proportionality to rainfall depends on the nature of the weather so that, say, twenty inches of rain in one part of the world may not give as much fallout as the same amount of rain in other weather zones. We are rapidly learning about this.