This paper was first presented at a symposium held at the University of British Columbia Faculty of Law on November 5 and 6, 2004 to honour the late Mr. Justice Kenneth Lysyk, a former faculty member and Dean of Law at U.B.C. For this paper we chose a topic that combines both of Ken Lysyk's favourite subjects. We set out to examine how the Supreme Court of Canada has used the "real and substantial connection" test in the conflict of laws and in related areas of constitutional law. This test has been adopted for a variety of purposes. We suggest that it serves some of these purposes better than others. In addition, we suggest that the test, as it is presently structured, serves none of its purposes especially well. The law makes frequent use of criteria that turn on an overall appreciation of a variety of factual elements. The law cannot do without such criteria, and often they function as the core concept for an area of law. The "real and substantial connection" test is such test (or actually, we would argue, several such tests using the same verbal formula for distinct purposes). We contend that the test does its job, or jobs, less well primarily because the standpoint from which the evaluation is performed is much less well-defined than it is in the others. A secondary reason is that the variety of contexts in which the test is employed makes its underlying rationale even harder to pin down.