4 SMITHSONIAN MISCELLANEOUS COLLECTIONS VOL. ^2 



ranging from half a minute to a minute. After each reading the 

 thermometer was shaken down to be ready for instant use when 

 needed again. 



Temperatures were read to tenths of a degree, and were tran- 

 scribed at once in a note book opposite the name of the bird, together 

 with the date and other needed information. In each case the time 

 of day was noted to the nearest quarter hour. Thus a temperature 

 taken at five minutes past ten in the morning was recorded as lo.oo 

 a. m. and one at ten minutes past ten as 10.15 a. m. Further refine- 

 ment in recording the time of day was considered useless, as save on 

 certain meridians there is always slight variation between the actual 

 time as shown by the height of the sun, and the time adopted for 

 universal local use as a matter of convenience. The time of observa- 

 tions recorded during the continuance of the so-called " daylight 

 saving" regulations of 1918 and 1919 has been corrected in each 

 instance to normal. 



In order to be prepared at all times to secure temperature records 

 of birds it was necessary to have thermometers constantly at hand 

 while in the field. The physical labor involved in field work at times 

 is arduous, so that though carrying cases of an improved type were 

 used many instruments were destroyed in spite of every precaution. 

 Some were broken through accident, others were crushed by birds 

 while records were being secured, and a few were lost. In all a 

 considerable number were used in securing the records given. 



The records of temperatures of birds secured have been taken 

 throughout the year, but the localities where this work has been 

 carried on have all been in the limits of the United States in temperate 

 regions where extreme cold in winter has not been encountered. As 

 a matter of fact very few of the observations have been made at a 

 time when the temperature of the air was below -f 25° Fahr. For 

 this reason it has not been necessary to adopt means for warming the 

 thermometers, or for keeping them warm, imm'ediately previous to 

 their use. During the early course of the observations recorded 

 herein experiments were made in heating thermometers by holding 

 them in the mouth of the observer when it seemed probably that they 

 were to be used in a short time. In this way the temperature of the 

 bulb and the glass for a short distance above was warmed to -1-98°, 

 more or less (depending upon the condition of the mouth). It was 

 soon found however that the results gained with warm and cold 

 thermometers were so nearly identical that it was impossible to dis- 

 tinguish between them. Where observations on the body tempera- 

 tures of birds are to be made at times when the air is at zero Fahren- 



