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

I've just finished reading some of my early papers, and you know, when I'd finished I said to myself, 'Rutherford, my boy, you used to be a damned clever fellow.' (1911)

If your experiment needs statistics, you ought to have done a better experiment.

I must confess it was very unexpected and I am very startled at my metamorphosis into a chemist.

[From uranium] there are present at least two distinct types of radiation one that is very readily absorbed, which will be termed for convenience the α radiation, and the other of a more penetrative character, which will be termed the β radiation.

It is essential for men of science to take an interest in the administration of their own affairs or else the professional civil servant will step in - and then the Lord help you.

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