to which PP and Rh belong, has found that the undisputed members of the tribe are all rather homogeneous in their wood anatomy.
this is in agreement whit the finding of this study. Also inagreement is the fact that Koek-Noorman’s observations differed in several repects from those of Metcalfe and Ghalk (1950), who stated that the members of the Rubiaceae wite non septate fibres typically have apotracheal parenchyma. This generalization does not hold true for the stem woods examined here, but the trunk wood of R.lujae does have apotracheal payrenchyma.
In the stem wood, the ray type and the structure of vessel pit vestures appear to be useful diagnostic characters. The rays of P.vollesenii and R.fischeri are the most similar,both having cells containing solitaly crystals,with sclerified ray cells being present only in the former.Crystals are absent from the ras of E. longiflora and F.aeruginescens. The inner vessel walls were examined with the SEM,and showed large differences between P.vollesenii and R.fisrcheri. In P.vollesenii there are very few vestures on the margins of the pit apertures,and the inner vessel walls are nor warted. In R.fischeri there are