3/o NOTES OF A BOTANIST CHAP. 



upon them. There are also cattle- farms ; and 

 around the wells from which water is drawn and 

 served to the cattle the same weeds are continually 

 springing up ; while the seeds, even of those that 

 grew on the desert, remain embedded in the sand 

 and retain their vitality during all the years of 

 drought. When the rains come on, therefore, they 

 cause, as it were, a unilateral development of the 

 vegetation from the forest across the open grounds, 

 and a corresponding expansion of the insect-life 

 which breeds and feeds upon it. 



Results the same in principle, but diverse in 

 mode, would take place under different local circum- 

 stances. Thus, if we suppose an oasis in the midst 

 of a desert exposed to the same exceptional access 

 of moisture as the desert of Chanduy with its forest 

 skirt, there would be generated an extension of 

 organic life radiating outwards in all directions. 



Besides the migrations above recorded, I have 

 many times in South America seen butterflies flying 

 across rivers so wide that it is impossible to suppose 

 they could be guided by any indication of sight or 

 smell. Animals of higher organisation and stronger 

 reasoning powers would probably turn aside along 

 the shore of the river or ocean in quest of food for 

 themselves and their offspring ; but there are 

 plainly cases where frail little creatures, such as 

 butterflies, must go straight forward at a venture, 

 and either attain their object or perish. 



MIGRATING ANTS 



The movements of Ants registered in my journal 

 are (as may be supposed) chiefly such as were 



