5o8 Bulletin 316 



earth. It is obvious, therefore, that the temperature of the surface on 

 which frost forms must be at or below the freezing point. Since ice forms 

 at a temperature of 32° F. and it is the formation of ice in the 

 intercellular spaces of a plant that results in injury, the temperature of 

 the plant thus becomes the determining factor, and not the appearance of 

 any deposit of frost on its surface. When frost forms on a plant, it is 

 good evidence, of course, that the exterior of the plant is at or below the 

 freezing point of water; but if the temperature at which condensation of 

 moisture in the air takes place is below freezing, as is frequently the case, 

 ice may form in the intercellular spaces of the plant and destroy it without 

 any deposit of frost appearing on its surface. For this reason, the term 

 " frost " as used in this bulletin is to be understood as signifying injurious 

 temperatures without regard to any deposit of ice that may or may not 

 have appeared on the exterior of plants or other objects. 



HOW OBSERVATIONS OF FROST ARE MADE 



It is the practice of observers connected with the United States Weather 

 Bureau to record three degrees of frost — light, heavy, and killing. Alight 

 frost is recorded when only the tenderest plants are injiired; a heavy 

 frost, when the injury isniore marked and the hardier plants are damaged ; 

 and a killing frost, when the staple products of the region are injured 

 severely or are killed. When no deposit of frost occurs and the tempera- 

 ture of the air as indicated by the thermometer falls during the night to 

 32° F., a killing frost is recorded also. The charts and tables 

 presented herewith are based on records of killing frosts only. 



The records of frosts made at the regular Weather Bureau stations in 

 the State — New York City, Albany , Canton, Syracuse, Oswego, Rochester, 

 Buffalo, Ithaca, and Binghamton- — represent to a considerable degree the 

 conditions with respect to frost that obtain in those cities. 



FROSTS IN CITY AND COUNTRY 



It is well known that the air over cities, particularly on clear nights 

 when frost is likely to occur, almost invariably is warmer than the air 

 over the open country. A comparison of the minimimi night tempera- 

 tures in the city of St. Louis, Mo., with those at Forest Park, a suburb 

 of St. Louis, for a period of five years shows that the air at night over the 

 city averaged 4.6° warmer than that over the suburb. For the 

 month of September the average difference at night was 9 . A 

 similar comparison of observations made since 1882 at the United States 

 Weather Bureau station in the city of Columbus, Ohio, and on the campus 

 of Ohio State University, three miles distant, shows that the air six feet 

 above ground at the university is 3.3° colder at night than that 



