B. TERRESTRIAL PHYSICS AND METEOROLOGY. 61 



for the year, and special memoirs relating to important mete- 

 orological phenomena. 



The Post-office bulletins are published daily at 4500 differ- 

 ent post-offices, through the co-operation of the War and Post- 

 office Departments. At the request of Professor Baird, Unit- 

 ed States Commissioner of Fish and Fisheries, observers fa- 

 vorably located are required to take daily observations of 

 the temperature of the water in the respective rivers and 

 harbors. In obedience to the Act of Congress of March, 1873, 

 a telegraph line has been completed along the coast of New 

 Jersey for a distance of fifty miles, and will rapidly be ex- 

 tended from Cape Hatteras to Cape Cod, Signal Service ob- 

 servers being stationed every ten or fifteen miles along the 

 coast. At the request of the Hon. Francis A. Walker, Super- 

 intendent of the United States Census, meteorological charts, 

 showing the average storm-frequency, the velocity of the wind, 

 and other interesting details, have been prepared to form part 

 of the statistical atlas of the United States. A laro-e number 

 of logs of vessels have been collected by the office, in pursuance 

 of the study of the phenomena of the storms experienced at 

 sea. One hundred and twenty-five extensive storms have been 

 experienced in the United States during the year, of which 88 

 have been sufficiently severe to demand the display of caution- 

 ary signals. The winds considered as dangerous by the Signal- 

 office, and justifying the display of the cautionary signal, are 

 those that reach a velocity of twenty-five miles per hour, as 

 registered by the anemometer on the land. The direction of 

 the wind is not taken into consideration. If within eio;ht 

 hours the danger appears not to be imminent, the signal is at 

 once ordered down. It is not displayed for an indefinitely 

 long time, as is the custom in Europe. Cautionary dispatch- 

 es are sent to Canada when storms apparently threaten Ca- 

 nadian ports. The percentage of verifications of the predic- 

 tions contained in the so-called probabilities varies from 82 

 in New England to 74 in the Northwest. These percentages 

 have been obtained by a careful analysis of the predictions, 

 and a comparison of them with the facts subsequently ob- 

 tained from the weather reports. The percentage of verifi- 

 cations exhibits a proportion to the number of stations from 

 which reports have been received, and thus, to a limited ex- 

 tent, demonstrates that the rules employed in the deductions 



