4o8 The Ohio Naturalist. [Vol. XV, No. 3, 



On the 15th the predicted temperature would be 34, while 

 the thermometer reached 33. On the 16th and 17th the exact 

 minimum would have been predicted. 



Rules to follow. The average time of this median hour will 

 vary under different weather conditions, at different seasons of 

 the year, and in different localities. Outside of the cities, in cen- 

 tral Ohio, under conditions of clear skies and comparatively still 

 air, it will be close to the following. 



April, 7:15 p. m. ; May and June, 7:30 p. m.; September, 

 6:30 p. m.; October and November, 6 p. m. In July it is about 

 7 :30 p. m., and in August, 7 p. m. 



If a strong wind is blowing in the afternoon or if the afternoon 

 is cloudy or partly cloudy, and the wind goes down and it clears 

 off in the night the time of the median temperature will be from 

 30 to 45 minutes later than the average given. 



If it should cloud up during the night after a clear afternoon 

 and evening the minimum temperature will not be quite so low 

 as is indicated by the median. 



In cloudy and stonny weather, or when strong southerly winds 

 prevail, or if the wind is high from the northwest the time of the 

 median varies so much that no attempt should be made to make 

 predictions from it. 



This is especially true when after a period of wann weather 

 the wind shifts to northwesterly and the temperature begins to 

 fall rapidly. This indicates the approach of a cool area and the 

 only way to estimate the probable minimum temperature is from 

 the daily weather maps. 



But after the windy front of this cool wave has passed by and 

 the air is clear and still and the days are warm and the nights 

 cool and frosts threaten then the plan can be used. 



Reliable maximum and minimum thermometers should be 

 obtained and exposed in a lattice work shelter where the air 

 circulates freely and the sun will not strike the instruments. 



The difference in temperature between that at the average 

 half-way or median hour should be subtracted from the highest 

 during the day and the difference subtracted from the reading at 

 the half way hour. The remainder will show the approximate 

 lowest temperature during the coming night. 



Records that are at hand indicate that the average time of 

 median will be slightly later in the valleys than at higher elevations, 

 but each man interested should be able to determine his own 

 median hour by careful records of the temperature. 



The Weather Bureau office at Columbus will continue its study 

 during the coming year at a larger number of stations than were 

 in operation last year. 



