The paper in which Albert Einstein proposed his light-quantum hypothesis
was the only one of his great papers of 1905 that he himself termed
``revolutionary.'' Contrary to widespread belief, Einstein did not
propose his light-quantum hypothesis ``to explain the photoelectric
effect.'' Instead, he based his argument for light quanta on the
statistical interpretation of the second law of thermodynamics, with the
photoelectric effect being only one of three phenomena that he offered
as possible experimental support for it. I will discuss Einstein's
light-quantum hypothesis of 1905 and his introduction of the
wave-particle duality in 1909 and then turn to the reception of his work
on light quanta by his contemporaries. We will examine the reasons that
prominent physicists advanced to reject Einstein's light-quantum
hypothesis in succeeding years. Those physicists included Robert A.
Millikan, even though he provided convincing experimental proof of the
validity of Einstein's equation of the photoelectric effect in 1915. The
turning point came after Arthur Holly Compton discovered the Compton
effect in late 1922, but even then Compton's discovery was contested
both on experimental and on theoretical grounds. Niels Bohr, in
particular, had never accepted the reality of light quanta and now, in
1924, proposed a theory, the Bohr-Kramers-Slater theory, which assumed
that energy and momentum were conserved only statistically in
microscopic interactions. Only after that theory was disproved
experimentally in 1925 was Einstein's revolutionary light-quantum
hypothesis generally accepted by physicists---a full two decades after
Einstein had proposed it.