Shadows are intriguing phenomena. They do not have mass or energy. So, they are unable to have some basic characteristics of the objects of which they are shadows: they cannot move by themselves and they cannot experience the same kind of changes. At first sight, any theory of perception can skip this optical phenomenon or look at it only as a side-effect. Actually, in order to be seen objects must be illuminated and one of the consequences of this is that they project a shadow over the surrounding space. Is that all? In this paper I will argue that, from a phenomenological point of view (or at least from a Husserlian oriented phenomenology), shadows, with their specific hyletic data, must be considered as an element of the process of constitution of spatial-temporal objectivities. In other words, shadows no less than other predicates, like extension or hardness, although in a different manner, belong to the