Under-boulder cryptic epifaunal micro-habitats in both the intertidal and sublittoral were modelled by means of 15 X 15 X 1 cm slate panels bolted together horizontally, but separated by perspex spacers of either 25,lO or 5 mm height. The under-surface of the top panel in each unit pair is held to represent a boulder undersurface and it is only these sessile assemblages which are considered. In order to assess the importance of timing of space availability on assemblage development, triplicate sub-sets of panels of all three gap-sizes were immersed in both habitats in February, June, August and October 1983. These were followed until JanuaryIFebruary 1986. Rates of larval settlement, organism growth and area occupied were consistently greater in the sublittoral, and showed more-or-less predictable patterns with gap size (25 > 10 > 5 mm). High-profile solitary organisms were not, as a rule, excluded from the narrower gap-sizes. Overall, for both habitats, colonial organisms dominated most panel sets, although solitary ascidians were occasionally important. The axiom of the solitary strategy dominating intertidal hard-substratum assemblages does not, therefore, extend to cryptic habitats. Observations of comparable panel-sets initiated in similar months in succeeding years and compared at 12 months' age showed marked differences in assemblage structure: within- and between-year stochasticity of larval settlement and establishment are clearly important in determining at least the initial stages of "succession". Qualitative and quantitative comparisons clearly showed the intertidal and sublittoral assemblages to be structurally quite different. Low-shore under-boulder cryptic epifaunal assemblages should not, therefore, be viewed as merely the uppermost fringes of otherwise sublittoral components.