Several particle physics theories predict the existence of suitable
candidates for composing the dark matter in the Universe. If these
particles originate from physics at the TeV scale, they could be
created at the Large Hadron Collider, and their annihilations might be
detected as radiation in gamma-ray telescopes. Alternatively, as
suggested by anomalous cosmic ray data, dark matter could be composed
by light hidden sector particles. These hidden sectors are probed in
high luminosity beam-dump experiments, supplemented by studies of
stellar evolution. I will review the constraints on different
candidate particles, and, using a method due to Eddington, discuss how
the velocity distribution can affect the prospects for indirect
detection of dark matter.