Though we understand the main steps which lead from diffuse gas to a newborn star with its nascent planets, a number of details remain obscur and need clarification. One of the least understood steps is the formation and collapse of the prestellar core, in the dark cloud. This is due to the core being extremely cold (down to 6 K), deeply embedded in the cloud and largely depleted of most molecular species. Therefore almost no tracer is available to study this phase in detail. Apart from dust itself (it would require a second talk to discuss it), the main tracers are H2D+, N2H+, N2D+ and DCO+ (the NH3 family is another possibility). 3 species out of 4 carry a Deuterium. Why are they abundant enough to be detected when the cosmic D/H ratio is as low as 10^-5 and what can we learn from their presence ? In the presentation, I will illustrate 3 different measurements we can perform thanks to the peculiar deuterium chemistry: the age of clouds, the age of prestellar cores and the depletion profile of CO and N2 in these cores, which is a prerequisite to study the interaction of grain surface chemistry (ices) with the gas phase in the dense cores. Deuterium appears as a swiss-knife to study star formation.