BY E. C. ANDREWS. 533 



In all these observations, the writer inferred that the predomi- 

 nant influence, in the plant-evolution, was geographical environ- 

 ment. Other factors in their evolution appeared to be Time, 

 Heredity, Variation, and Selection. Time is a constantly flowing 

 quantity, and, in the evolution of floras, such an evenly flowing 

 quantity must be large, since the other factors produce only in- 

 finitesimal results when acting only during brief periods. The 

 remaining factors are exceedingly variable in their action, and the 

 sluggish or rapid response of the organism, along variable lines, is 

 due mainly, either to the fixity of its climatic and soil-environment, 

 or to a revolution in its geographical surroundings. A geographi- 

 cal revolution would tend to produce either plant-extinction in 

 part or as a whole, or a relatively sudden deployment into new 

 genera or even families, such as one sees among the Dicotyledons 

 in later Mesozoic time. 



The principle adopted, in the present note, is the application 

 of the "Law of Probabilities" to the case of the development and 

 distribution of the Myrtaceae Thus the earlier types of the family 

 have been sought by a consideration of those genera which possess 

 the most points in common, consonant with certain facts known to 

 be connected with the geographical setting of the earlier types. In 

 this way, those types are considered as aberrant which, although 

 excellently adapted to a definite set of local geographical condi- 

 tions, nevertheless depart in essentials from the deduced primitive 

 forms. Thus, for example, if it should be ascertained that the 

 points common to all genera of the Myrtaceae were much more 

 characteristic of the genera not endemic to Australia, such as the 

 Myrteae ; furthermore, if such Myrteae were ascertained to be prac- 

 tically confined to the tropics; that they flourished in good soil and 

 abundant rainfall; that their species far outnumbered those of 

 types endemic to Australia; furthermore, if it were found that 

 Australia had been isolated from the other parts of the tropics 

 during the production of the endemic genera; that the Australian 

 types flourished on porous, sandy soil, and in proportion as they 

 tended to depauperate types, that they exhibited modifications of 

 those organs typically developed in the Myrteae, then these would 



