Marine snow plays a key role in the global carbon cycle because it transfers huge amounts of carbon dioxide (around 1-2 Gt per year) from the ocean surface to the deep-sea, thus removing it from the global system. It is of major field of study for several decades to quantify the amount of particulate matter settling through the water column. A central parameter for ocean mass flux estimates is the settling velocity of larger particles. Most of the few available datasets have been acquired by Scuba divers but are they are limited to a diving depth of a few tenth of meters. Particle settling speeds for the deeper water column may be estimated with the help of sediment trap recordings, having the disadvantage to integrate settling speeds over a long period of time and for the entire particle population settling through the water column. In situ sinking speed measurements of individual aggregates however, are rare and difficult to obtain. We present results from a settling chamber constructed for in situ sinking speed measurements of marine snow. The settling chamber was mounted to the MARUM Cherokee ROV during RV Poseidon Cruise 365 in 2008 off Cape Blanc, Mauritania. It was constructed in consideration of a similar device used by the MBARI ROV Ventana. It is a simple plexiglas box which can be opened and closed to allow an infinite number of measurements with little disturbance inside. A collimated light source illuminates a defined sample volume in which aggregates can be observed after the box has been closed. We sampled a total of 51 aggregates at four depth levels, between 50m and 400m water depth. The depths were chosen after collecting a vertical particle profile acquired by a deep-sea still image camera system before the deployment of the ROV. Sinking speeds ranged from 10m d-1to 287 m d-1 with a mean value of 57 m d-1. No clear relationsship between the size of the particles and their sinking speed was found. Furthermore we could not observe increasing particle sinking speeds with increasing water depth as found by other authors. This underlines the complexity of such studies and implies more deployments during upcoming cruises and comparison of in situ measurements with additional methods in the future.