360 Kershaw . — The Structure and Development of 
central strand in the nucellus of Myrica Gale , which is a comparable 
structure to the tracheids in Castanea as also f a vestige of some long lost 
structure \ 
It is interesting to find that in two families generally regarded as 
closely related to the Myricaceae, the structure of the ovule resembles 
very closely that of Myrica Gale. 
Nicolofif 1 in his treatise on the Juglandaceae produces several figures 
of the ovule of Juglans regia which show that very similar features are 
found to those which have been described for Myrica Gale. In a longi- 
tudinal section of the ovule which is orthotropous, he figures the nucellus 
as being quite free to the base from the integument (Nicoloff, 1 . c., Fig. 18). 
In a transverse section he represents the integument as having a ring of 
vascular bundles, ten in number (Nicoloff. 1 . c., Fig, 31). Although these 
characters are described by Nicoloff he does not suggest in any way 
their probable phylogenetic value. 
In the recently investigated allied order of Julianiaceae, 2 the ovule 
of Juliania is described as having an integument containing a branching 
system of vascular bundles. The nucellus is figured as being fused to 
the single integument however. This may be accounted for by the fact 
that the ovule is hemi-anatropous, not orthotropous as in the cases described 
where the nucellus is free. 
The absence of these two apparently primitive characters of the 
ovule in the allied groups of Amentiferae and Casuarinaceae may be con- 
nected with the change from the orthotropous to the anatropous type of 
ovule. 
A comparison of the structure of the ovule in the families Juglandaceae, 
Julianiaceae, and Myricaceae may throw some light on the position of these 
families in the Natural Classification, a point which has given rise to some 
discussion. 
Hallier 3 regards the Amentiferae including Myricaceae as a degenerate 
group derived probably from the Terebinthaceae in which are included 
Anacardiaceae, Juglandaceae, and Julianiaceae. 
Hemsley, 4 on the other hand, regards the groups Juglandaceae, 
Julianiaceae, and Amentiferae as very closely allied, the Julianiaceae 
probably occupying a position immediately between Juglandaceae and 
Amentiferae. 
The evidence derived from the structure of the ovule undoubtedly 
1 Nicoloff: Sur le type floral et le developpement du fruit des Juglandees. Journal de 
Botanique, t. xxviii-xxix. 
2 W. Botting Hemsley : On the Julianiaceae. A New Natural Order of Plants. Phil. Trans. 
Roy. Soc., vol. cxcix, 1908. 
3 Hallier : ‘ Uber Juliania , eine Terebinthaceen-Gattung mit Cupula, und die wahren Stamm- 
eltern der Katzchenbliitler : Dresden, 1908, 
4 Hemsley : loc. cit. 
