112 VARIATION IN IvEPTlNOTARSA. 



Strongly contrasted with the species just discussed are forms like L. unde- 

 cimlineata and dahlhomi. UndecimUneata, while variable, does not show 

 even in its wide range of habitat any great geographical variability, nor has 

 it been possible in experiment to produce any extreme modifications. L. dahl- 

 bomi, of all species in the genus, has been found in the most varied habitats — 

 in the hot, humid valleys of Guatemala and southern Mexico, in the dry, cold 

 plateau, in the dry, hot Rio Grande Basin, in Yucatan, and in other diverse 

 places, yet in all it is the same conservative species, and as far as known is 

 almost free from variation of any kind. 



If the data of geographical variation were to be presented for the different 

 elements of the color pattern, for size, variation in punctation, and for gland- 

 ular openings, it would but be a repetition of what has just been said concern- 

 ing general color variation in these beetles, and nothing would be added to 

 the value of the above discussion. 



It appears that geographical and place variation are closely related; that 

 place variation is in reality the cause of geographical variation. Of most 

 importance, however, is the fact that, as far as it has been possible to deter- 

 mine, geographical variations are transient and form no permanent part of the 

 species. They are, as it were, a suit of clothing used as the weather dictates, 

 and produce no more permanent constitutional modification than does the act 

 of changing one's dress. 



Some authors, as, for example, Allen, have attempted to establish a rela- 

 tionship between latitude and longitude and variation. As far as this material 

 is concerned, any such relationship is purely incidental and of no importance. 

 Correspondence between geographical variation and any artificial scheme for 

 dividing the earth's surface by arbitrary lines would, on the employing of 

 enough imagination, show a relationship. The only real relations of geo- 

 graphical variation that I can discover are those to natural features of the 

 earth's surface, to climate, topography, and other natural phenomena. 



We have already seen the relation which exists between place variation and 

 geographical variation, and I have shown that place variation is due to the 

 varying conditions of different environmental complexes. In the tables given 

 are found examples of the most exact correspondence of variation to differ- 

 ent portions of the habitat. These relations will come out better and in their 

 true light in the next section of this paper. 



