io88 
Thompson . — The Anatomy and 
Fig. 25. But a single vessel is often sufficient to show clearly the gradual dis- 
appearance of the torus and border. In PI. XCV, Fig. 20, at the end of the 
vessel the pits have a typical border and a deeply stained, well-marked 
torus. Towards the bottom of the field, i. e. towards the centre of the end 
wall of the vessel, the torus becomes gradually fainter until it disappears, 
and there results an open communication through the pits. The decrease 
in breadth of the border is equally gradual. Usually the conditions are not 
so diagrammatic as this. Sometimes the torus and border do not disappear 
at the same rate, but the latter may be quite well developed when the former 
has completely vanished. Diagrammatic cases are sufficiently numerous, 
however, and the various stages presented by different elements are so 
often repeated that there can be no doubt as to the origin of the perforations, 
in Ephedra at least. 
In those cases in which the torus is only partly developed, it may have 
become so in one of two ways : either in the course of ontogeny it has 
ceased to grow at that stage in its development, or it may have continued 
to the complete condition, and later have been partially resorbed. The 
writer was unable to determine which of these alternatives is correct. 
If the vessels have developed in the manner indicated it might be 
expected that the seedling would present primitive conditions approaching 
tracheides. As a matter of fact the seedling possesses very few vessels of any 
kind, the mass of wood consisting almost exclusively of tracheides. This 
fact is shown strikingly in PI. XCVI, Fig. 27, which is a photograph of 
a transverse section through the base of the seedling stem of E. altissima. 
Very few vessels can be recognized, and even those which can are extremely 
small. The absence of vessels is also shown in longitudinal view in Fig. 28. 
Of course, as one examines the upper parts of the seedling it is found that 
they gradually increase in number. Moreover, the vessels which are present 
have only a single row of perforations, and these are usually of the transitional 
type. Such facts furnish another striking example of the recapitulation in 
the seedling of ancestral characteristics. 
It should also be noted that in the first annual ring of adult branches 
the few vessels which are present usually have a single row of perforations 
which are often transitional to bordered pits. Evidence is thus added to 
that which has recently been accumulated showing that this region is also 
the seat of retention of ancestral features. 
Boodle and Worsdell 1 state that the node also lacks vessels. For the 
species which the writer has examined this statement does not appear to 
hold good, although the number in that region may be slightly less than in 
the internode. The presence of vessels in the node of E. altissima is shown 
by photography in Pl. XCIV, Fig. 7, especially at the right. 
We have here seen the structure of the primitive vessels of Ephedra 
1 loc. cit. 
