140 T O R R E Y A 



to which the species is adapted there will be, for it, a progressive deterioria- 

 tion of the climatic type. That is, in marginal regions of the climatic type where 

 it begins to grade into another climatic type, there will be fewer and fewer 

 suitable spots for the species. Of necessity, if this picture be true, the density 

 of the species will tend to decrease toward the periphery. Some interesting data 

 concerning the behavior of species at the margin of range have been published 

 by Griggs (1914) on the Sugar Grove district of southern Ohio. He says, "It 

 is clear . . . that in this region the species in which the individuals become 

 scarcer and scarcer until it fails altogether is exceptional." Certain species are 

 approximately continuous up to the margins of their range, but others are in- 

 creasingly discontinuous until they are characteristically disjunct, and some- 

 times widely so, in the peripheral portion of their areas. 



In the light of these data, it would seem that the criterion of species domi- 

 nance and density is by no means an infallible guide to center of origin. Domi- 

 nance and density are frequently highly irrelevant in this respect. 



Criterion 3. Location of Synthetic or Closely Related Forms 



From the context and through correspondence I find that by "synthetic" is 

 meant generalized or primitive forms of a phyletic group. With this half of the 

 criterion we can have no quarrel this far : the most primitive form or forms of 

 a group certainly arose somewhere, and wherever that was, there is the center 

 of origin of the group. To ascertain that center, after a group has had a long 

 history, is, however, another matter. 



It is frequently claimed that the center of origin for a group is where the 

 earliest fossil forms have been found, whether or not the group is represented 

 there today. For example, it has been claimed that the shell family Pleuro- 

 ceridae had a western origin because its earliest record is from the Laramie 

 formation (Colorado, etc.). Adams (1915), however, concluded that the fam- 

 ily, and especially lo, had a southeastern origin centering in eastern Tennessee 

 despite the absence of substantiating fossils. 



There are two diametrically opposed views. The most widely accepted 

 view is that the most primitive members of a group are still to be found at or 

 near the center of origin of the group. This is frequently true because most of 

 our temperate genera date back to the Cretaceous or early Tertiary and their 

 primitive forms are frequently found concentrated in the old land areas. In 

 the United States, for example, such ancient land masses with primitive spe- 

 cies (Gleason, 1923) include the Southern Appalachian center, the Cumber- 

 land and Ozark center, the prairie center of Nebraska, Kansas, and eastern 

 Colorado, the southwestern desert center, etc. In a study of Lesquerclla, Payson 

 (1922) concluded that the center of origin of the genus was in the old land 

 area of central Texas where "not only are these species primitive, but in no 



