The Abietopitys petriellae woods here described have primitive gymnospermous characters, such as narrow rays, alternate pitting and crowded continuous bordered pits that completely fill the radial wall. These features are already present in some Devonian fossil woods (Beck, 1970).
The genus Abietopitys is characterized by the peculiar thickenings in the walls of ray cells. This type of thickenings is unique in the Paleozoic woods, although some woods referred to the genus Callixylon have ray tracheids with secondary thickenings that slightly resemble these thickenings. The Abietopitys genus was originally recognized for the Permian Ecca Group of South Africa (Kräusel, in Kräusel and Range, 1928; 1956), and then two fragments of secondary xylem from the Permian of Chubut province, Argentina, were assigned to the genus by Archangelsky (1960). Therefore, based on the present results the range of the genus is extended to the Upper Carboniferous in sediments from Argentina.
The leaf traces observed, only two in a new specimen, and three in the holotype, are insufficient to make a mapping. Unfortunately, in the other transverse sections, no leaf traces were found.
Because fossil woods are not usually found in organic connection with other parts such as leaves or reproductive organs, the botanical affinity of most Paleozoic woods, such as those described in this paper, is difficult to determine. However, it was possible to compare with the plants of known affinity present in the same beds. Among them, Cordaites and Pteridospermales are recognized.
The axes are similar in the pitting, among other features, to woods of Cordaites, although the pith is not septated, as in most northern Cordaites. Much less is known about the southern Cordaites, and an affinity to this group cannot be discarded. It is also different from other typical pteridosperm stems in having only one stele. In addition, they could also be related to primitive pteridosperms with one stele like Eristophyton Zalessky, (see Galtier and Scott, 1990, for a description of Eristophyton and comparisons with other Paleozoic gymnosperms).
In any case, the presence of the same association of leaves and fertile structures with woods in different localities should be taken into account to infer the probable affinity of the woods.
The author thanks S.N. Césari and P.R. Gutiérrez for collecting the new material and S.N. Césari and J. Garcia Massini for reading the manuscript. He is also grateful to J. Galtier and B. Meyer-Berthaud for their valuable comments as reviewers. The funding for this work was provided by the Agencia Nacional de Promoción Científica y Tecnológica (PICT 0482).