It is widely accepted and almost certainly true that both pulsars and supernova remnants (SNRs) are products of the collapse of a star at the end of its evolution. Given this, it is a considerable puzzle why, of the more than 120 known SNRs in the Galaxy, only two have unambiguously associated pulsars. Beaming of the pulsar emission probably accounts for the absence of detectable pulsars in up to 80% of the SNRs; however, this still leaves 20–30 SNRs in which one should be able to detect a pulsar. Vivekanand and Narayan (1981) show that there is a deficit of pulsars with periods ≲0.5 s and suggest that a majority of pulsars do not become active for a time ∼104 years after their birth. This would account for the lack of pulsar-SNR associations. It is however possible that the observed lack of short-period pulsars is simply due to observational selection. In the past, most pulsar searches have been made at relatively low radio frequencies, typically close to 400 MHz. At these frequencies SNRs are bright and the effects of interstellar scattering are significant, especially for distant, short-period pulsars. Further, most of these searches have used a relatively long sampling interval, typically about 20 ms, which further reduces the sensitivity for short-period pulsars.