Evolution and development of the arthropod head are heavily debated topics often referred to as "The arthropod head problem" (AHP) or the "Endless dispute". One aspect of the AHP concerns the evolutionary origin and homology of the most anterior head segments, the pre-gnathal segments (PGS), that are associated with the tripartite brain of arthropods. It has been suggested that the PGS could have evolved independently from the segments that build the rest of the arthropod body. One argument supporting this hypothesis is that the PGS are patterned by an initial transverse stripe of hedgehog (hh)-expression that splits once or twice, giving (in the case of double splitting) raise to the three PGS in distantly related arthropods such as the fly Drosophila melanogaster and the spider Parasteatoda tepidariorum. It has been implied that this splitting-event may recapitulate evolution of these segments, i.e. the splitting of an initial anterior body unit into three, or at least that the single splitting may represent a remnant of this process. In this paper, I show that two-fold splitting of an initial anterior hh-stripe is not conserved in chelicerates or even spiders. Instead, I find that a single splitting event correlated to the development of the most anterior two segments, the protocerebral and the deuterocerebral segment, is conserved among arthropods as a whole. There are, however, deviations from this pattern including a third or even fourth consecutive head segment, or even hh-splitting in more posterior segments.