One of the most fundamental unknowns about pulsars is the mechanism bywhich they radiate. Although many steps have been taken towards understandingthis in the 48 years since their discovery, there is still much to learn. The recentlaunch of the Fermi γ-ray satellite has led to an astonishing increase in thenumber of pulsars detected in γ-rays, making this an ideal time to consider theemission at both low energies (radio) and high energies (γ-rays).In this thesis we use radio polarisation data and a novel method to determine the viewing geometries of two samples of pulsars, one consisting of youngpulsars which have been detected in γ-rays and the other consisting of comparable young pulsars which are not detected in γ-rays. We find that the magneticinclination angle distribution of the γ-ray-detected pulsars does not match thecommonly-expected random distribution for such young pulsars as these, insteadbeing skewed towards lower values. We find two possible explanations for this.The first is that the expectation of a random distribution is erroneous, a situationwhich will have important consequences for pulsar population studies, considerations of the origins of magnetic fields in neutron stars and models of the supernovaprocess. The alternative is that the radio beams are larger than expected by afactor which depends on the magnetic inclination, which would have implicationsfor the magnetic field structure.We also find that the γ-ray- and non-γ-ray-detected pulsars differ in radioprofile width. This leads us to present a unified model of these young pulsarswhere the γ-ray detectability is highly dependent on magnetic inclination, whichdetermines if our line of sight samples the γ-ray emission. We report for the firsttime ever a significant correlation between the radio and γ-ray profile morphology. This is independent confirmation that the existing models of the radio andγ-ray emission regions are, at least qualitatively, realistic, although significantquantitative problems are reported.