186 



RECORD AND RESULTS OF 



Supposed Dependence of the Winter Temperature on the Lunar Phases. 



The supposed lower temperature about the time of full moon when compared 

 with that about new moon, during mid-winter, noticed by some Arctic explorers, 

 and which received contirmation from observations during two winters at Van 

 Rensselaer Harbor, and partial confirmation from observations during two winters 

 in Baffin Bay and at Port Kennedy, is not sustained by the observations at Port 

 Foulke, as may be seen from the following collection of mean daily temperatures, 

 each the mean of five days, two of which precede and two of which follow the lunar 

 phase ; to allow for the annual change of temperature the alternate means are set 

 out. These alternate mean temperatures, and the observed temperatures, are then 

 compared by subtracting the temperature at the new moon from that at full moon ; 

 a negative siajn indicates 2:reater cold at full than at new moon. 





Observed 



Alternate 



Difference 





temperature. 



means. 



o— o 



O October 29, 1860 



—o°.n 







• November 13, " 













+ 4.5 



+ 80.4 



+30.9 



O November 28, " 













+ 17.5 



—7.2 



+ 24.7 



• December 12, " 













—19.0 



—0.2 



+ 18.8 



O December 28, " 













—18.0 



—18.4 



+ 0.4 



• January 11, 1861 













—17.8 



—23.2 



—5.4 



O January 26, " 













—28.5 



—21.7 



—6.8 



9 February 9, " 













—25.7 



—24.8 



+ 0.9 



O February 25, " 













—21.2 



—21.6 



+ 0.4 



• March 11, " 













—17.6 



—21.2 



—3.6 



O March 26, " 













—21.3 



—17.9 



—3.4 



• April 10, " 













—18.2 



—13.8 



+ 4.4 



O April 24, " 













—6.2 



+ 5.1 



—11.3 



• May 9, " 













+ 28.4 







If we take the differences from the middle of December to the end of March, 

 the temperature would appear 2°. 5 colder at full than at new moon; the high tem- 

 perature about November 28, and the low temperature about December 12, how- 

 ever, are such strong contradictions to the supposed law, as to deprive the results 

 collected by the expedition of any decisive value. About November 28, the pre- 

 vailing wind was S. W., charged with heat and vapor from the open water spaces 

 of North Baffin Bay; about December 12, the prevailing wind was N. E. Neither 

 Port Foulke nor Port Kennedy are favorably situated for the experimental study 

 of the phenomenon. 



Relation of tlie Atmospheric Temperature to the Direction of the Wind. 



The method pursued to ascertain the elevating or the depressing infiuence of the 

 various winds on the temperature of the air, is as follows : The average daily tem- 

 perature for each day of the year was computed by means of the expression for T, 

 this was readily done by the use of the formula for a number of equi-distant inter- 

 vals, and by the application of the principle of intei^polation " into the middle" 

 (which secures the proper value to third differences inclusive). The previously used 

 correction for graduation of thermometers was next applied loith sign reversed so as 



