Usually, only great apes are supposed to be able to recognise themselves in front of a mirror, but so far, only few other primate species have been tested for this ability. In particular, marmosets as "lower" monkeys are not expected to pass the mark test as an indicator for self-recognition. In fact, our test animal Callithrix jacchus showed no signs of any kind of self-recognition when confronted with a mirror for the first time. However, we started from the assumption that self-recognition is not a all-or-nothing phenomenon that appeared all of a sudden in certain species and not in others, but that it has evolved step by step in the phylogeny of primates. Therefore, the aim of our experiments was not only to test the marmosets for spontaneous self-recognition, but to identify possible precursors of this cognitive capability as well. So we tried to improve the animals´ understanding of their own reflection by different tasks they had to solve, as for example identifying the mirror as an object in itself or using the mirror instrumentally to localise and grasp at visible and hidden food respectively. In addition, we tested the presence of object permanence to show that the animals are able to imagine, at least to a certain degree, invisible objects bearing in mind that such a capacity should be an important prerequisite to correctly understand the reflections of a mirror. Finally, by discreetly spreading some chocolate cream on the animals´ forehead, we carried out a new variation of the classical mark test to enhance the motivational interest in the mirror. Motivation, then, changed in fact dramatically and the observed behavior more precisely unveiled the cognitive limits of Callithrix jacchus. Further tests are planned to check if these limits possess already a definitive, i.e. (phylo)genetic character or if there still exist some possibilities to evade them by a variation of certain external stimuli.