Based on the mapping of non-redundant known M. incognita subventral (SvG) and dorsal (DG) gland effectors on the genomes of five Meloidogyne species ancestral numbers were deduced across their phylogeny using parsimony inference with Mesquite (Maddison & Maddison, 2014). In the genomes of M. incognita, M. arenaria, M. javanica and M. entetolobii, only genes mapping with >90% identity on >66% of their length were considered unambiguously present (character state ‘1’ in Mesquite). Genes with a significant spaln mapping but that did not meet the % identity and % coverage thresholds were considered as unsure to be present (character state ‘?’ in Mesquite). Because M. hapla is phylogenetically much more distant, only a length coverage threshold was used to infer unambiguous presence following a significant spaln mapping from the ATG position. Genes that returned no significant spaln mapping were considered absent from the genomes under consideration (character state ‘0’ in Mesquite). The tree topology of the five Meloidogyne species used for ancestral states reconstruction was taken from (Álvarez-Ortega, Brito, & Subbotin, 2019). We used the analysis ‘Trace all Characters’ to reconstruct characters (presence / absence of each effector) at each ancestral branch. To infer the minimal unambiguous number of effectors at each ancestral branch the sum of 1’s at this branch was calculated while equi-parsimonious 0/1’s ancestral state were considered an absence in a conservative scenario.