This study was designed to establish whether populations of fish species important in the catches of rock and surf anglers increased following the proclamation of the De Hoop Marine Reserve on the southern coast of South Africa. Catch per unit effort (CPUE) and size frequency distributions of angling species were monitored approximately monthly by angling from the shore at 2 sites in the reserve. At one site sampling commenced 2 yr before the reserve was established and continued 4.5 yr thereafter, allowing comparisons of periods of exploitation and protection. The other site had a long history of minimal exploitation and data collected there 2.5 to 4.5 yr after complete protection was considered to represent an unexploited condition. Ten species accounted for 99% of the catch. Following establishment of the reserve, CPUE increased for 6 of these (galjoen Coracinus capensis, dassie Diplodus sargus, wildeperd D. cervinus, white steenbras Lithognathus lithognathus, Cape stumpnose Rhabdosargus holubi and musselcracker Sparodon durbanensis). The catch rates of 2 species (C. capensis and D. sargus) improved 4 to 5 fold within 2 yr of protection and remained at these high levels, which were similar to the unexploited catch rates. Recoveries were slower amongst the other 4 species, their catch rates reaching 30 to 60% of the unexploited level after 2.5 to 4.5 yr of protection. This study provides evidence of general stock recoveries of exploited fish species in a shallow marine habitat following protection within a marine reserve.