E. V. Wulff — 36 — Historical Plant Geography 



complicated is the problem of determining the location of the center 

 of an area, and that for its solution statistical calculations alone can 

 have only a very limited significance. It is necessary to assemble data 

 of an all-sided study of species— as, for instance, was done by Ko- 

 MAROV (1908)— disclosing the phylogenetically most primitive types, 

 the direction of their evolution, the centers of concentration of these 

 primitive, initial types, and the direction of their further distribution. 

 Only such a monographic study, based also on paleobotanic data, can 

 give a more or less correct idea as to the initial center of the area of a 

 genus and of the secondary centers of its development. 



In this respect cytological data may prove of great value. It has 

 now been established that in some cases species belonging to the same 

 genus differ in chromosome number, and that many of the polyploid 

 species originated, apparently, as a result of chromosome mutations 

 induced by the action of external factors. Species, in dispersing from 

 the center of their origin, often extend their area beyond the boundaries 

 of optimum conditions for their existence. As a result of the action of 

 ecological factors to which a species is not accustomed there occur 

 irregularities at meiosis in the sex cells, which result in the phenomenon 

 of polyploidy. This connection between the origin of polyploid species 

 and definite ecological conditions is the reason why such species have 

 in many cases quite specific geographical areas, differing from the areas 

 of the initial species (see below; also, in more detail, Wulff, 1937). 



It is now possible, therefore, theoretically to advance the proposition 

 that floras of those regions of the globe characterized by extremely low 

 temperature, such as arctic regions and mountain peaks, or by very 

 high temperature and low humidity, such as deserts, are distinguished 

 by an exceptionally large number of polyploid species. It must also be 

 presumed that such chromosome mutations have occurred in nature 

 not only under present-day conditions, as a result of species having 

 become widely distributed and having penetrated into localities with 

 ecological conditions differing from those normal for them, but also as a 

 result of the great climatic changes that took place in former geological 

 times and of the migrations of species in those times. 



Arranging the species of a genus in order according to chromosome 

 number, we obtain so-called polyploid series of species, which at the 

 same time reflect the directfon of evolution of the genus and also the 

 direction of its dispersal. Starting from the premise that the species 

 having the smallest chromosome number in a polyploid series usually 

 is the initial species, we may consider that the areas of species charac- 

 terized by such chromosome numbers are more ancient than areas of 

 species with larger chromosome numbers and that, consequently, in the 

 regions occupied by these ancient areas one must seek for the initial 

 center of the area of a genus. As an illustration we may cite the data 

 of a cytological investigation of the genus Iris carried out by Simonet 

 (1932). This genus is widely distributed throughout the entire north- 

 ern hemisphere. Species having rhizomes occupy the largest areas, 

 practically identical to that of the genus, while tuber-bearing species 

 are considerably more restricted in their distribution. The area of the 

 latter is confined to the Mediterranean Basin— from the Iberian Penin- 

 sula to Soviet Central Asia, inclusive. Not only all four sections 



