In the original self-test, in which 1,000 runs were conducted with a Random Number Generator (each run being 100 trials long), a comparison was made between the 46 runs in which I reported the presence of sensations of Kundalini and those 954 runs in which Kundalini was absent. The mean deviation score for Kundalini runs was above chance to a degree approaching significance; the mean for non-Kundalini runs was at chance; and the mean for the Kundalini runs was higher than that for non-Kundalini runs to a degree approaching significance. It was concluded that these results merited follow-up, and there were in fact 4 replications. In Replication 1, the mean for Kundalini runs was below ...