,3,,. A NEW GROUP OF FLOWERING PLANTS. 143 



Believing that he has clearly demonstrated that Casuarina con- 

 stitutes the type, and probably the only representative, of a separate 

 class, Treub proposes the following division of Angiosperms : — 



Sub-group Angiosperms. 

 Sub-Division. Sub-Division. 



Chalazogames. Porogames. 



Class. . Classes. 



Chalazogames. Monocotyledons ; Dicotyledons. 



This grouping is open to criticism, and we cannot but think that 

 the author has selected the least important difference to characterise 

 his new subdivision ; for the route taken by the pollen-tube is surely 

 of little account provided the macrospore is ultimately reached. 

 Until we have some more definite evidence of the relation in time 

 between impregnation and the formation of the endosperm, we cannot 

 lay much stress upon it for purposes of classification. It is quite 

 possible that fertilisation never occurs, and that the oosphere born with 

 its cell-wall already formed is potentially an oospore. There are 

 cases of apogamy quite as striking in the fungi among the Sapro- 

 legnieai. The fact that nuclei were never distinctly seen in the apex 

 of the pollen-tube points in this direction, as also does the compara- 

 tively thick membrane of the oosphere, which would afford consider- 

 able resistance to the passage of fertilising material. The most 

 striking fact is the number of the macrospores, and the high develop- 

 ment they attain. Rosa livida, referred to above, shows an indication 

 of the same, though in a much less degree. 



Dr. Treub has certainly made out some points of the highest 

 interest, but hardly sufficient, especially considering the perfect 

 similarity of the embryology with the ordinary dicotyledonous type, 

 to separate the genus Casuavina as a distinct subdivision of flowering 

 plants. 



A. B. Rendle. 



