A loss of large vertebrates has occurred in aquatic and terrestrial ecosystems, but data to measure long-term population changes are sparse. Historical photographs provide visual and quantitative evidence of changes in mean individual size and species composition for groups of marine fish that have been targeted by sport fishing. I measured such trends for 13 groups of recreationally caught "trophy" reef fish with photographs taken in Key West, Florida, from 1956 to 2007. The mean fish size declined from an estimated 19.9 kg (SE 1.5) to 2.3 kg (SE 0.3), and there was a major shift in species composition. Landings from 1956 to 1960 were dominated by large groupers (Epinephelus spp.), and other large predatory fish were commonly caught, including sharks with an average length of just <2 m. In contrast, landings in 2007 were composed of small snappers (Lutjanus spp. and Ocyurus chrysurus) with an average length of 34.4 cm (SE 0.62), and the average length of sharks declined by more than 50% over 50 years. Major declines in the size of fish caught were not reflected in the price of fishing trips, so customers paid the same amount for a less-valuable product. Historical photographs provide a window into a more pristine coral reef ecosystem that existed a half a century ago and lend support to current observations that unfished reef communities are able to support large numbers of large-bodied fish.

The productivity and biomass of pristine coral reef ecosystems is poorly understood, particularly in the Caribbean where communities have been impacted by overfishing and multiple other stressors over centuries. Using historical data on the spatial distribution and abundance of the extinct Caribbean monk seal (Monachus tropicalis), this study reconstructs the population size, structure and ecological role of this once common predator within coral reef communities, and provides evidence that historical reefs supported biomasses of fishes and invertebrates up to six times greater than those found on typical modern Caribbean reefs. An estimated 233000-338000 monk seals were distributed among 13 colonies across the Caribbean. The biomass of reef fishes and invertebrates required to support historical seal populations was 732-1018gm-2 of reefs, which exceeds that found on any Caribbean reef today and is comparable with those measured in remote Pacific reefs. Quantitative estimates of historically dense monk seal colonies and their consumption rates on pristine reefs provide concrete data on the magnitude of decline in animal biomass on Caribbean coral reefs. Realistic reconstruction of these past ecosystems is critical to understanding the profound and long-lasting effect of human hunting on the functioning of coral reef ecosystems.

Degradation of coral reef ecosystems began centuries ago, but there is no global summary of the magnitude of change. We compiled records, extending back thousands of years, of the status and trends of seven major guilds of carnivores, herbivores, and architectural species from 14 regions. Large animals declined before small animals and architectural species, and Atlantic reefs declined before reefs in the Red Sea and Australia, but the trajectories of decline were markedly similar worldwide. All reefs were substantially degraded long before outbreaks of coral disease and bleaching. Regardless of these new threats, reefs will not survive without immediate protection from human exploitation over large spatial scales.