This review presents results obtained in recent years concerning the catalytic conversion of methane into higher hydrocarbons using metal catalysts under non-oxidative conditions at moderate temperature. Although only a limited amount of work has been carried out in this area, the non-oxidative homologation of methane has already proved itself to be a novel and interesting way of addressing the problem of methane upgrading. Chemisorption of methane on transition metal surfaces has been studied for long on either ill-defined surfaces or single crystals. These studies, reviewed here, concerned mainly the kinetics and dynamics of chemisorption, In contrast, little has been known concerning the nature and reactivity of the adspecies. The reactivity of C-1 adspecies originating from precursors other than methane (carbon monoxide, diazomethane, ketene, etc.) is better understood, which can be useful in investigating the reactivity of the hydrocarbonaceous adspecies resulting from methane. Most of the work concerning homologation of methane under non-oxidative conditions has been done by the groups of Amariglio in France and van Santen in the Netherlands. Both use two-step procedures in which metal catalysts are exposed first to methane and then to hydrogen. However, the procedures differ markedly in essentially two points: (i) the temperature of the first step and (ii) the pressure of methane. The Dutch group always uses a two-temperature cycle, decomposing dilute methane on Ru and Co at a rather elevated temperature and then carrying out hydrogenation at a much lower temperature and at atmospheric pressure, The French group, in contrast, has shown that homologation can be performed isothermally and at a moderate temperature on Pt, Ru and Co, using methane and hydrogen at atmospheric pressure. Consequently, in the two procedures the nature and reactivity of the surface species formed at the end of the exposure step are different. When the exposure to methane has been carried out at a moderate temperature and at atmospheric pressure, C-gamma is not formed (therefore no irreversible poisoning) and the products do not obey the Anderson-Schultz-Flory distribution. Also, thermodynamic limitations are circumvented by both groups but for reasons specific to each procedure. Finally, this review outlines prospects for future research and attempts briefly to estimate the potential commercial interest of the concept.