24 REPORT OF THE SECRETARY. 



highest mean monthly temperature seems to occur almost exactly in 

 the middle of July, and the lowest point would probably have been 

 found in February if the series had been extended over several 

 winters. The mean temperature of winter, namely, of December, 

 January, and February, was — 28°. 59; of spring, — 10°. 59; sum- 

 mer, -(-33°. 38; autumn, — 4°. 03. The mean temperature for the 

 whole year Avas — 2°. 46. The temperature was always lowest. 

 during calms, and rose with the springing up of a wind from any 

 quarter. 



There is also a great regularity in the elevation of temperature 

 during the hours of the fall of snow; on an average the sensible heat 

 was increased during this period 7°. 7. In seventeen months it 

 snowed during six hundred and eighty hours, and rained during sixty 

 hours. 



A series of recurring periods of cold was observed, which Dr. Kane 

 seemed inclined to consider as intimately connected with the phases 

 of the moon, and on this point a series of elaborate investigations- 

 was made by Mr. Schott, from Avhich it was found that in a period 

 of six days on an average the cycle was completed, and that the 

 lowest temperatures are reached about the time of full moon. Setting 

 aside some small deviations in the regularity of the curves of tem- 

 perature, there is not a single exception to the correspondence of the 

 greatest cold near the epoch of full moon, and of least cold near the 

 time of the new moon. It should be observed, however, that since, 

 from the observations made at this Institution, the waves, as it were, 

 of cold air which reduce the temperature of the United States, 

 frequently begin several days earlier at the extreme west, the same 

 coincidence as to identity of occurrence of the maximum cold with 

 any particular phase of the moon cannot be true of all points on the 

 surface of the earth, although the period of recurrence may, as in the 

 case of the tides at different places, be governed by that luminary. 



A series of comparative observations at the level of the sea and at 

 the top of the mast of the brig, at eighty feet elevation, was taken 

 during the months of August, September, and October, from which 

 is deduced a diminution of temperature of 1° for two hundred and 

 ten feet of elevation. 



The direction of the wind was noted in the original records with 

 reference to the magnetic points of the compass, and the mean results 



