2:iO E. J. BttTLEft. 



a distance of eight miles. From the same place or from Madiki 

 infection spread to Rajahmundry, 13 or 10 miles respectively, in 

 about two years. On the other hand, Vileasavilli is said to have 

 been first attacked in 1906, while Amalapur town, 4 miles distant, 

 was certainly infected before 1900. So also an observer noted that 

 the village of Rajathapalapatnam was still free from disease in 1907, 

 while Vetlapalam, an adjoining village,' had it since 1902. This 

 sharp limitation of the disease at the border of the infected area has 

 been several times noted. Near Kolanka, which was almost cer- 

 tainly attacked before 1900, the disease in one direction observed 

 in 1907 diminished from severe (over 50 per cent, of deaths) to little 

 (under 5 per cent.) in two miles, and two miles further on none could 

 be found. Such cases as the last connect with those in which the 

 onflow of infection appears to have received a check which shows 

 some promise of being permanent. These are chiefly to be found 

 towards the sea and are associated witli sandy soils, but there is 

 also some evidence that the drier upland tracts above the Delta 

 oppose a barrier which may perhaps be found efficient. There is 

 no doubt that there are at least two entirely distinct sets of factors 

 controlling the extension. The first is meteorological and, other 

 things being equal, the rate of spread will largely depend on the 

 amount of rain, fog, dew and similar conditions. This will be more 

 fully considered below. The second is biological and includes the 

 little understood conditions which determine the virulence of the 

 parasite, and still more, the susceptibility to attack of the victim. 

 These seem to be in the present case closely connected with the 

 influence of the nature of the soil on the palm itself. 



Though, in a broad sense, the disease occupies a continuous 

 area, it must not be supposed that it is uniformly distributed w^ithin 

 this area. On the contrary, there is the greatest difference in the 

 severity of the attack from village to village and even from field 

 to field. Close observation confirms the view that the nature of the 



1 It should be noted here that the term " village " is used in the sense employed by the 

 revenue authorities, and denotes a certain area of land belonging cither to one large village or 

 to several hamlets. The first affected portion of the one village might therefore be perhaps 

 a couple of miles from the nearest part of the adjoining unaffected village. 



