HIGHWAYS OF DISPERSAL. 4 1 



lished : Hamilton County, Ohio, 1868 ; Lancaster County, Pennsylvania, 

 1872; and Charleston, South Carolina (1892), and Savannah, Georgia (1885) 

 (plate 8). 



None of these, however, changed in any material way the general move- 

 ment. The seven trends of migration found all follow natural highways of 

 dispersal or paths of least resistance. These highways are absolutely inde- 

 pendent of organisms, but are the results of geologic and geographic forces 

 which are at work upon the continent. In the main these highways are 

 associated with river valley developments, and also, to some extent, with 

 coastal plains. The existence of these natural highways has been recognized 

 by many writers, but Webster (1898, 1903), who has studied the trends of 

 insect migration, does not seem to have correlated them in any degree with 

 natural topographic highways. 



L. deccmlineata, on its escape from its habitat, entered an area devoid of 

 natural barriers, across which it was able to spread in any direction, with 

 the results shown on plate 8, where the yearly lines of advance show a 

 nearly even expansion in all directions. What irregularities there were 

 were produced by winds and moisture, which we shall discuss presently. 

 As soon as it reached the Mississippi Valley, the first great natural highway 

 that was encountered, there was a marked change in its behavior. Previous 

 to 1863-64 it had crossed wide stretches of country each year, but after 

 entering the Mississippi Valley its advance was checked for two years. In 

 this period we find it extending northward with great rapidity, reaching 

 nearly to the headwaters of the river in 1865, and also southward, although 

 much more slowly, down the valley. Thus this natural highway served 

 for movement in both directions at the same time, but much more rapidly 

 in one direction than in the other. Webster (1898, 1903) has thought of 

 the Mississippi Valley as a highway only for northward migration of southern 

 forms, but Adams (1902) and others regard it as a pathway for both north 

 and south migration, and it is here shown that movement in both directions 

 occurs. The difference in the rate of movement in the two directions is due 

 largely to winds, temperature, and moisture, climatic factors whose directive 

 influence I shall discuss later. 



On passing eastward out of the Mississippi Valley the beetle again spread 

 out, this time in a broad wave over Illinois, Wisconsin, and western Indiana. 

 This wave extended from near the head of the Mississippi Valley southward 

 to the junction of the Mississippi and Ohio rivers and eastward to the 

 southern end of Lake Michigan. Over this territory natural barriers of any 

 size or importance were absent. On passing the southern end of Lake 

 Michigan it entered the great natural highway of the Great Lakes-St. 

 Lawrence system. Along this it now spread with great rapidity in the years 

 1866 to 1880 northeastward across the State of Michigan and through 



