The internal anatomy is often well preserved, and there is a number of species of leaves whose anatomy is known. As will be expected from the parallel veins, in each section there are many vascular bundles running equidistantly through the tissue. Fig. 62A shows the microscopic details from a well-preserved leaf. In all the species patches of sclerenchyma were developed, and everything indicates that they were tough and well protected against loss of water, even to a greater extent than are most of the leaves of living Gymnosperms. In the stems the pith was much larger than that in living Gymnosperms (where the wood is generally very solid), and it was hollow in older stems, except for discs of tissue across the cavity. The internal cast from these stems has been described before, and is seen in fig. 63. [Illustration: Fig. 62B.—Much-magnified Wood Elements from _Cordaites_ Stem seen in longitudinal section, the type known as _Araucarioxylon_. Note the hexagonal outlines of the bordered pits, which lie in several rows] The wood was formed in closely packed radiating rows by a normal cambium (see p. 66), and the tracheæ so formed had characteristic rows of bordered pits (see fig. 62B). The wood comes nearer to that of the living Araucarias than any other, and indeed the numerous pieces of fossil wood of this type which are known from all the geological periods are called _Araucarioxylon_.[11] A double strand goes out from the main mass of wood, which afterwards divides and subdivides to provide the numerous bundles of the leaf. [Illustration: Fig. 63.—Cast of Hollow Pith of _Cordaites_, the constrictions corresponding to discs of solid tissue across the cavity] In the case of these fossils we are fortunate enough to have the fructifications, both male and female, in a good state of preservation. As in other Gymnosperms, the male and female cones are separate, but they differed less from each other in their arrangement than do those of any of the living types hitherto mentioned. They can hardly be described as true cones, though they had something of that nature; the seeds seem to be borne on special short stems, round which are also sterile scales. In the seed and the way it is borne perhaps the Cordaiteæ may be compared more nearly with the Taxeæ than with the other groups. A seed, not yet ripe, is shown in slightly diagrammatic form in fig. 64, where the essential details are illustrated. The seeds of this family sometimes reached a considerable size, and had a fleshy layer which was thick in comparison with the stone, and externally comparable with a cherry—though, of course, of very different nature in reality, for _Cordaites_, like _Taxus_, is a Gymnosperm, with simple naked seeds, while a cherry is the fruit of an Angiosperm.