Spur blight (Didymella applanata), cane Botrytis (B. cinerea) and L. coniothyrium were studied in 2 consecutive seasons in unsprayed raspberries and in plots sprayed with thiophanate-methyl in the first season alone, either twice preharvest, twice postharvest or twice pre- plus twice postharvest. Preharvest but not postharvest sprays gave moderate control of the first 2 diseases which occurred mainly on the lower halves of canes, B. cinerea being the less common. L. coniothyrium occurred at the base of canes where they were wounded by old cane stubs. In the first year infection was severe; 37% of canes died before harvest in the unsprayed plots. The 3 spray programmes all decreased cane death due to L. coniothyrium and increased yield by c. 45% even in those plots sprayed postharvest, when spur blight was severe and not controlled. This last disease clearly had no effect on potential yield. In the second year cane blight was common but less severe and lesion incidence was reduced similarly by all programmes, but only 4% of canes died in control plots. Analysis of potential yield and L. coniothyrium lesion size, in canes inoculated at fortnightly intervals in the previous year, showed that potential yield loss occurred only when lesions girdled canes. A bimodal distribution in the lesion sizes indicated some endogenous control of lesion development which might explain the marked differences in the effect of cane blight on potential yield in 2 seasons.