Origin of Biomes and Succession 287 



less humid conditions than those required by their parent species. 

 After the Paleocene these plants colonized progressively more xeric 

 sites in southwestern North America (Fig. 61) and evolved many 

 adaptations to life under xeric conditions such as water-storage, 

 seasonal deciduousness, and extremely short life in annuals. The 

 present product of this series of events is the live oak, pifion- 

 juniper, and other woodland associations of southwestern North 

 America, and their respective successional stage communities. 



In reconstructing the evolution of these and other biotas (Axel- 

 rod, 1958; Chaney, 1947; MacGinitie, 1953) no attempt was made 

 to trace the evolutionary patterns of individual successional com- 

 munities. Such a segregation would indeed be impossible from 

 fossil evidence alone because not only are successional stages, in- 

 cluding the climax, usually mixed intimately in the same area and 

 hence contributing jointly to local fossil deposition, but also pre- 

 climax successional stages are short-lived temporally and in the 

 geologic record would not be found separately except in varve 

 deposits. However, in both MacGinitie's and Axelrod's studies the 

 fossil species match so closely the composition of some living suc- 

 cessional communities that the mid-Cenozoic semi-desert plants 

 could very well have been organized in the same community struc- 

 ture we find at present. The fossil evidence is comprised chiefly 

 of woody species which are therefore representative of later suc- 

 cessional communities. Preclimax shrub communities are repre- 

 sented by the genera Ceanothus, Cercocarpus, and Arctostaphylos. 

 Undoubtedly the earlier and chiefly herbaceous communities also 

 existed but Axelrod mentions that herbaceous types were "largely 

 unrecorded." 



Information concerning the behavior of existing succession com- 

 bined with fossil evidence of studied floras supports three simple 

 principles concerning the community colonization of abiotic areas: 



( 1 ) Members of the first stage subclimax community, the normal 

 bare ground invaders, would be the first organisms to extend their 

 range onto previously uninhabited ground. Only after these had 

 become established would species of the next community in the 

 normal succession have any opportunity to extend their ranges into 

 the newly colonized areas and in essence follow the ecological ex- 

 tension of the first colonizing community. This process would be 

 repeated for each community in the succession. Based on the as- 

 sumption that later subclimax stages in the parent area cannot 

 initiate growth on bare ground within that area, it is unreasonable 



