SUMMARY:The two central paradigms of marine diversity are that there is a latitudinal cline of increasing species richness from poles to tropics and that species richness increases with depth to a maximum around 2,000 m and thereafter decreases. However, these paradigms were based on data collected in the late 1950's and early 1960's. Here I show that the 1960's data, are not representative and thus the paradigms need re-examination. New data from coastal areas in the northern hemisphere record species richness as high as the highest recorded in the deep-sea. Whilst this suggests that the cline of increasing diversity from shallow to deep-sea does not exist, however, the database for the deep sea is not sufficient to draw such a conclusion. The basic problem with the data from the 1960s is that samples were taken on ecological scales and yet they are used to answer evolutionary questions. The questions that such data were to answer were why do the tropics have higher species richness than polar regions or why do deep-sea sediments have more species than coastal sediments? Evolutionary questions need data from much larger spatial areas. Recently, data representative of large scales have been collected from coastal areas in the northern hemisphere and show that there is a cline of increasing species richness from the Arctic to the tropics, but there does not yet seem to be a similar cline in the southern hemisphere. A number of hypotheses have been proposed for the observed patterns in biodiversity. In terrestrial ecology the energy-productivity hypothesis has gained wide acceptance as an explanation for the latitudinal gradient. Here I examine this and other hypotheses critically. Finally an analysis of research priorities is made. Assessment is urgently needed of the spatial scales and dynamics of species richness from point samples to assemblages, habitats and landscapes, especially in coastal areas and in the tropics, where the threats to biodiversity are greatest. New technologies are available, such as side-scan sonar, acoustics, and under-water digital video cameras but as yet have been relatively little used. Conservation of marine biodiversity must be based on sound theory, yet marine diversity studies lag well behind those of terrestrial habitats.