A 104-Ma record of deep-sea Atelostomata (Holasterioda, Spatangoida, irregular echinoids) - a story of persistence, food availability and a big bang.

Deep-sea macrobenthic body fossils are scarce due to the lack of deep-sea sedimentary archives in onshore settings. Therefore, hypothesized migrations of shallow shelf taxa into the deep-sea after phases of mass extinction (onshore-offshore pattern in the literature) due to anoxic events is not constrained by the fossil record. To resolve this conundrum, we investigated 1,475 deep-sea sediment samples from the Atlantic, Pacific and Southern oceans (water depth ranging from 200 to 4,700 m), providing 41,460 spine fragments of the crown group Atelostomata (Holasteroida, Spatangoida). We show that the scarce fossil record of deep-sea echinoids is in fact a methodological artefact because it is limited by the almost exclusive use of onshore fossil archives. Our data advocate for a continuous record of deep-sea Atelostomata back to at least 104 Ma (late early Cretaceous), and literature records suggest even an older age (115 Ma). A gradual increase of different spine tip morphologies from the Albian to the Maastrichtian is observed. A subsequent, abrupt reduction in spine size and the loss of morphological inventory in the lowermost Paleogene is interpreted to be an expression of the "Lilliput Effect", related to nourishment depletion on the sea floor in the course of the Cretaceous-Paleogene (K-Pg) Boundary Event. The recovery from this event lasted at least 5 Ma, and post-K-Pg Boundary Event assemblages progress-without any further morphological breaks-towards the assemblages observed in modern deep-sea environments. Because atelostomate spine morphology is often species-specific, the variations in spine tip morphology trough time would indicate species changes taking place in the deep-sea. This observation is, therefore, interpreted to result from in-situ evolution in the deep-sea and not from onshore-offshore migrations. The calculation of the "atelostomate spine accumulation rate" (ASAR) reveals low values in pre-Campanian times, possibly related to high remineralization rates of organic matter in the water column in the course of the mid-Cretaceous Thermal Maximum and its aftermath. A Maastrichtian cooling pulse marks the irreversible onset of fluctuating but generally higher atelostomate biomass that continues throughout the Cenozoic.