20 REPORT OF THE SECRETARY. 



7j degrees. This difference is less than that deduced from the ob- 

 servations of Kane at Van Rensselaer harbor, but of the same kind, 

 and serves to establish the fact of the occurrence of maximum cold at 

 the period of full moon. Mr. Schott, in his paper on the discussions 

 of Kane's observations, has referred this cold to a secondary action. 

 The moon, as is now established by direct experiment, radiates and 

 reflects a considerable amount of heat to the earth, which is of the 

 kind called dark heat, or that of such low intensity as to be readily 

 absorbed by the vapor of our atmosphere, and particularly by clouds. 

 The effect, therefore, of the full moon is to dissolve the clouds, and 

 thus to give freer passage to the radiated heat from the earth into 

 celestial space. 



In order, however, that this explanation should be true, it is 

 necessary that the heat from the moon should be more penetrating 

 and have more effect upon the clouds than that from the earth; and 

 that this is the case is not improbable, as a part of the heat from the 

 moon is that reflected from the full meridian sun, while that given off 

 from the earth is merely due to its own nocturnal radiation. 



It is not probable, as we have stated in a previous report, that an 

 equal difference of temperature at the time of new and full moon will 

 be observed in middle latitudes, for, from the observations made at 

 this Institution, the waves, as it were, of cold which reduce the tem- 

 perature of the United States, frequently begin several days earlier 

 at the extreme west; and hence while the full moon occurs nearly at 

 the same moment of absolute time at all places on the surface of the 

 earth, the maximum cold might occur in one place at the new, and in 

 another at the full moon. In the arctic regions, on the other hand, 

 where the moon is at the same moment visible from every meridian, 

 the effects of its heat must be more perceptible and less masked by 

 the operation of other causes. In the observations of a long series 

 of years, however, the difference may, perhaps, be rendered manifest, 

 even in the latitude of Washington. 



Professor Dove, of Berlin, has called attention to the remarkable 

 recurrence of cold about the 11th of May of each year, but nothing 

 of this kind can be deduced from the observations during the voyage 

 of the "Fox," although from the observations by Kane at Van Rens- 

 selaer harbor on the 13th of May, 1854, the temperature was 9°. 3 

 lower than that computed for the mean of the same day at the same 

 place. 



The diminution of temperature at this period is evident from the 



