ANATOMY AND PHYLOGENETIC POSITION OF BETULACEAE 417 
many free nuclei is developed which after fertilization becomes a 
nutritive tissue. The fertilization is chalazogamic." These remarks 
according to Moss (1912) are based upon Treub's work. Moss, how- 
ever, goes on to say that Frye (1903) has shown Treub to be wrong 
in regard to the anomalous embryo-sac, large number of nuclei, etc. 
He remarks that all of these characters suggested by Treub have 
been found in Carpinus and many of them in Corylus, Betula and 
Alnus. These are such as chalazogamy (fertilization of the egg- 
nucleus through the chalaza rather than direct as in the case of the 
majority of the Angiosperms), the occurrence of more than one embryo- 
sac, the formation of a coecum from the originally isodiametric embryo- 
sac, the occurrence of tracheides among the embryo-sacs, the entrance 
of the pollen tube at the base of the coecum in Corylus and Carpinus, 
the closure of the micropyle, and finally the fusion of the ovule with 
the wall of the ovary. Frye's results are corroborated by the work of 
Benson (1894), who showed that chalazogamy not only prevails among 
the Casuarinaceae but also among the Betulae and Coryleae, and that 
Castanea as well as Casuarina possesses tracheides around the base 
of the embryo-sac. Again, Benson (1906), in collaboration with 
Sanday and Berridge in 1906, found still further evidence that Casua- 
rina differs in no essential from others of the Amentiferae. All genera, 
as she remarks, contain arborescent, wind-pollinated species. All are 
monoecious with flowers closely aggregated in unisexual catkins. The 
female flower in all is dimerous, bearing free stigmas, and the ovary 
of all genera is inserted in a radial plane. Casuarina differs from 
Carpinus exactly as does Quercus from Fagus, and the absence of a 
perianth in the female flpwer of Casuarina is a character common 
to Betula and Alnus. She states that the most important features 
of difference are its switch-like character and its phyllotaxy. These 
combined with some minor differences in floral structure and in em- 
bryology may suffice as grounds for forming a distinct group within 
the Betulaceae equivalent to the Coryleae. Thus he would dispense 
with the term Verticillatae and leave out such a cohort in the classi- 
fication of the Dicotyledons. 
From the above brief statement the reader may observe that the 
actual phylogenetic position of the Casuarinaceae remains at present 
somewhat in doubt. Although, as stated above, it is now placed in a 
cohort of its own at the very base of the Dicotyledons, yet there are 
recent data to show that it is really in no important respect distinct 
