A good scientific theory should be explicable to a barmaid.

If your result needs a statistician then you should design a better experiment.

I am a great believer in the simplicity of things and as you probably know I am inclined to hang on to broad & simple ideas like grim death until evidence is too strong for my tenacity.

Should a young scientist working with me come to me after two years of such work and ask me what to do next, I would advise him to get out of science. After two years of work, if a man does not know what to do next, he will never make a real scientist.

I congratulate you both on a fine piece of work which I am sure will ultimately prove of importance. I am personally very much interested in your results... In the past I have tried a number of experiments... but without any success.

Every good laboratory consists of first rate men working in great harmony to insure the progress of science; but down at the end of the hall is an unsociable, wrong-headed fellow working on unprofitable lines, and in his hands lies the hope of discovery.

The more physics you have the less engineering you need.