THE FALSIFIABILITY OP THE LORENTZ-FITZGERALD CONTRACTION HYPOTHESIS

IN his latest book The Logic of Scientific Discovery, K. R. Popper cites the LorenoF*ify tfcr -̂ n̂ confTfl ctTon "V iyMhf m ts ^^ frfl^iplc QJ JJJ nrrfjf t%* if tor y ftoxiltt^nr hypothesis', riaiming that i t ' had no falrifiable cnmrqtimirs but merely served to restore the agreement bctwuji theory and experiment—mainly thr tm<1ingt of Michebon and Morley '.* The charge of being ad hoc levelled here by Professor Popper against die LorentzHtzgerald contraction hypodiesis is wholly incorrect but has persisted in tbe literature on dbe philosophy of science for the past half century. I therefore wish to give a refutation of die theriiof non-fakifiabiliryon die basis of a companion of die '"""wg underlying die Kennedy-Thomdike experiment • with die design of die MkhebonModey experiment. Specifically, I shall show that, contrary to Professor Popper's contention, (i) die coupling of die Lorentz-Fitzgerald contraction hypothesis with die «-1«w-il aether theory entails a positive outcome of die Kennedy-Thomdike experiment differing quantitatively from die positive result required by die aether dieory iWtfamt die LorenoFitzgerald auxiliary hypothesis, and (ii) while wi failing the null outcome produced by the Michelson-MorUy experiment, die Lorentz-Fitzgerald version of die aedier dieory rules out die negative result which was actually yielded by die Kennedy-Thomdike experiment. The ry««ia1 difference between die apparatus used in these two experiments is the following: as measured by rods in die laboratory, die horizontal and vertical arms of me Mkhelson interferometer used in the Keimedy-Thomdike experiment are not equal but are made as different in length as possible, so as to assure a considerable difference in the travel times of die two partial beams from die source to die point at which they recomhinr to produce interference fringes. On die other hand, the horizontal and vertical arms of die apparatus used in the Michelson-MorUy experiment are each of die same length / as measured by die rods in die laboratory. Accordingly, in die hitter experiment die classically-expected round-trip times T, and Tk for die vertical and horizontal arms are given respectively by