510 Eighth Annual Report of the 



values can, however, be determined for a few points in various 

 sections of the State and the normals of adjacent stations may 

 be derived from these by comparing the shorter with the longer 

 records month by month, since the differences between the tem- 

 peratures at adjacent stations are much more constant than the 

 temperatures themselves. But before using this method it is 

 necessary to ascertain the limits of 'distance within which the 

 temperatures have a similar variation, especially in a region sit- 

 uated, like New York, near the usual path of storms. For ex- 

 ample, a cyclonic depression passing south of this State may 

 give a warm wave over the southern counties, while anticyclonic 

 conditions and extreme oold prevail in the northern section. 

 Further, the inversion of temperature, or decrease with height, 

 occurring within anticyclonic areas, is frequently the source of 

 deviation from the usual thermal relations between highland 

 and valley stations. 



