Those were just the early lucky ARPA fundees, mostly because they had excellent computer science departments (or were well-connected politically, like Harvard)

Those were just the early lucky ARPA fundees, mostly because they had excellent computer science departments (or were well-connected politically, like Harvard) or were research institutes like SRI or Rand. Plus the usual military nodes early on, since all the ARPA grants were sponsored by the military, and they had to keep their fingers on things.