Sykes— The Anatomy and Morphology of Tmesipteris. 83 
wards to enter the sporangiophore ’. This arrangement certainly suggests 
that c their present axillary position is secondary, not primary \ 
Finally, in the Sphenophyllales and Psilotales there are no longer any 
signs of the separate origin of bract and sporangiophore, and their close 
association has often been regarded as a matter of course. It appears to 
me that Lignier’s series 1 ( Sphenophyllum Dawsoni } Palaeostachya , Calamo - 
stachys ), which was constructed to illustrate the gradual shifting of the 
sporangiophore from the sporophyll on to the axis of the cone, may with 
equal ease be read in the inverse order, and then illustrates the removal of 
the sporangiophore from the axis on to the sporophyll. 
The fossil cone, Spencer ites 2 , perhaps throws some light on the possible 
origin of the Lycopods by reduction from some more complicated form. 
The proximal part of the sporophyll in Spencerites may represent the 
remains of a branch axis bearing a single bract and a sporangium on 
a short stalk. Scott 3 considers it likely that the present Lycopods are 
reduced rather than primitive, and Strasburger 4 suggested in 1873 that the 
formation of axillary bulbils, in a position exactly similar to that of 
sporangia, in Lycopodium Selago , is an argument for the theory that 
the sporangium in this genus is of the nature of a reduced axillary shoot. 
It seems quite possible that the axis of the original sporangial branch has 
become much reduced in the Lycopods ; according to this view a whole 
fertile branch in Tmesipteris is homologous with a single Lycopodian 
sporophyll. In conclusion, while, on this view, it is not possible to found 
a relationship between the Psilotales and Sphenophyllales on the ground 
of the common foliar nature of their sporophylls, it appears not unlikely 
that the primitively axial nature of the spore-bearing organs may yet be 
demonstrated in Sphenophyllum and Cheirostrobus , and that a relationship 
between the two families may be finally established. Each so-called 
‘ sporophyll ’ in the Sphenophyllales may perhaps be proved to be equivalent 
to a reduced branch bearing sporangia and leaves, the original axis of 
which has become much reduced during the process of formation of a cone 
from a lax arrangement of scattered fertile branches, such as is characteristic 
of the stems of Psilotaceae to-day. 
VI. Systematic Position. 
The fact that the gametophyte of the Psilotaceae is unknown is a great 
hindrance in the formation of any conception of their real relationship to 
other families. The only case of a prothallus possibly belonging to 
Psilotum was recorded by Lang 6 , and resembles a Lycopod prothallus of 
the ‘ clavatum ’ type. 
The Psilotaceae were formerly considered to be nearly allied to 
1 Lignier, 1903, p. 95. 2 Berridge, 1905. 3 Scott, 1907, p. 175. 
4 Strasburger, 1873. 5 Lang, 1904. 
