Figure 2

Non-metric multidimentional scaling (NMDS) plots representing community similarity (UniFrac distances) of bacteria (A,C) and eukaryotes (B,D) by host plant species identity (A,B) and by pH (C,D). Each point represents a sample and distance between points represents degree of similarity in community composition. Lines in A and B connect points belonging to the same species. Grey contour lines in C and D are smooth surfaces calculated based on variation in pH using the ‘ordisurf’ function in the ‘vegan’ package of R. Species codes: BOS = N. boschiana, COP = N. copelandii, EYM = N. eymae, FUS = N. fusca, INE = N. inermis, JAC = N. jacquelineae, KHA = N. khasiana, MAX = N. maxima, RAM = N. ramispina, SAN = N. sanguinea, SIN = N. singalana, TEN = N. tentaculata, TRU = N. truncata. pH levels: 1 = red, 2 = orange, 3 = yellow, 4 = green, 5 = blue, 6 = purple.