The ophidism represents a serious public-health issue in Brazil both due to their high occurrence and due to their morbidity/mortality as well. According to the Brazilian Health Ministry 19,000 to 22,000 snake envenom accidents occur each year. Most of these accidents are attributed to Bothrops species. This fact makes the taxonomical, immunological and venomic studies of this group extremely important. The taxonomical positioning of the species traditionally included in the Bothrops genus has been considerably modified in the past years. Such modifications have been based mostly on morphological characters, and on some biochemical traits (electrophoretic pattern of proteins from liver, kidney, heart, esqueletic muscle and blood plasma). In the present study, high performance liquid chromatography coupled to mass spectrometry (LC/MS) has been used to analyze the crude venom from 11 Bothrops species (Bothrops jararaca, B. jararacussu, B. neuwiedi, B. taeniatus, B. brazili, B. erythromelas, B. moojeni, B. cotiara, B. atrox, B. leucurus e B. castenaldi) in order to establish both taxonomical and phylogenetic relations between them. The large number of molecules found in these venoms has been clustered according to their physico-chemical properties (molecular mass and hidrophobicity), using the machine learning-based Weka software. In the assigned clusters the presence of some toxin structural families, such as BPPs, disintegrins, serineproteases and metalloproteases, could be verified. Then, a phenetic correlation tree has been generated from the clusterization process data, and was compared to existing taxonomical trees. In this way, we believe that this methodology is appropriated to determine the Bothrops venoms` complexities and to assess their taxonomical relationships as well.