Peter, you are missing one very crucial aspect of the scientific method: **peer review**. In no particular order, science is observation, hypothesis and peer review. Peer review is especially important to the former in the way that a [verifiable][1] hypothesis must also be [falsifiable][2]: if no one reviews your hypothesis and record of observed data, how can anyone verify your results? Consider that knowledge is empirical verification of what is (else how do you know what is?) and that for your record of observations and hypothesis to be verified as useful for making predictions, other people must review and scrutinize the hypothesis and experiments with their own observations and critique.

For example, as your peer and while reviewing your experiment of cutting off leaves I am motivated to question several aspects of what is otherwise an example of taking a hunch, making observations, suggesting a hypothesis and proposing a test to confirm or refute that hypothesis. Does this test the change of color from aging, or just from decay? How would you conduct an experiment to establish the cause of color changes during the life cycle of the plant and its leaves? How will you distinguish cause from correlation? What evidence might show that your hypothesis is false, or are you simply confirming what you want to be the case? Is it enough to make a general observation that certain species of trees have significant changes in the color of their leaves during the Autumn? Is your theory that tree leaves change colors related to decay anything more than a mundane statement of the obvious? Can you think of an abstract way of formulating this process?


  [1]: http://s-f-walker.org.uk/pubsebooks/pdfs/ayerLTL.pdf
  [2]: http://www.stephenjaygould.org/ctrl/popper_falsification.html