142 T O R R E Y A 



Texas. Also, T. hirsutiflora and T. ozarkana exhibit the same tendency. The 

 combination of cytology with geological history and taxonomy suggests very 

 strongly that the Edwards Plateau area of central Texas was the immediate 

 center from which the American tradescantias have developed in compara- 

 tively recent times. 



With respect to the other point of the criterion, it can be said that closely 

 related forms can come to be located almost anywhere within the generic area. 

 The nearest relative of any form, however, will tend to be near by, at least at 

 first, because of the filial relationship between them. According to Kinsey 

 (1936), the picture of evolution is that of a simple or infrequently branching 

 chain. In this chain each species is a derivative of a previously existing species, 

 usually without extermination of the parental species. 



When one looks at a large family of plants, it is apparent that it is not 

 everywhere equally well developed or rich. A certain tribe composing, say, 10 

 per cent of the family, may in one region constitute 30 or 40 per cent or more 

 of the family. This phenomenon is likely true for the other tribes. Such re- 

 gions of differentiation are likely regions of speciation or origin, except where, 

 for historical reasons, they are known to be regions of preservation. I can not 

 see, however, that closeness of relationship among species can ever be employed 

 as a criterion to indicate the geographic center of origin of a group without the 

 aid of other facts. We can only say that primitive and closely related forms may 

 or may not be at the center of origin. 



Criterion 4. Location of Maximum Size of Individuals 



In a discussion of evolution of species through climatic conditions, Allen 

 (1905) reiteratds some "laws" stated by him in 1882: (1) the maximum 

 physical development of the individual is attained where the conditions of en- 

 vironment are most favorable to the life of the species; (2) the largest species 

 of a group (genus, sub-family, or family) are found where the group to which 

 they severally belong reaches its highest development, or where it has what 

 may be termed its center of distribution. 



These conclusions were reached from the observation that "in the northern 

 hemisphere, in nearly all types of both birds and mammals of obviously north- 

 ern origin, there is a gradual decrease in the general size from the north south- 

 ward in the representatives of a conspecific group ..." Later on he says, "The 

 variation in size from north southward is as gradual and continuous as the 

 transition in climatic conditions." 



It seems to me that within these statements, employed by Adams and others, 

 the "cat is out of the bag." In the first place, size is a specific character that 

 may not be related to environment. Size differences may be due to biotype 

 selection across a climatic gradient, or to phenotypic expression. Allen's state- 



