How many atoms are in 99.7 grams of iron?

Explanation:

The correct answer is #1: The scientist collaborates with others to plan and carry out an experiment and then writes a paper that has a clear conclusion. He submits it to a professional journal for peer review, after which the results are presented publicly.

#2 seems less realistic, though there are cases when results of a secret study are shared with the world: the project is kept secret until the team reaches its goals to prevent others from using intermediate results to achieve the same goal faster. The method of sharing seems wrong though: the results should be published (rather than communicated directly to other scientists) to prevent the possible disputes regarding authorship.

#3 "lone genius" scenarios are rather rare. More importantly, the standard process is publishing results in a peer reviewed medium rather than calling a press conference immediately.

#4 is how science should work in the ideal world, but in reality collecting data from multiple diverse sources in hard, though publishing results in the Internet (rather than specialized journals) becomes more and more common.