204 ANNUAL EEPOET SMITHSONIAN INSTITUTION, 1915. 



In such a case the loss to the steamship companies in interest and 

 depreciation on ships and cargoes and in wages may easily amount 

 to more than $50,000 per day, and this loss occurs not once but fre- 

 quently during a year, and on many routes. 



In addition to this, the danger of collision in fog adds very con- 

 siderably to the cost of insurance, and some of our worst disasters 

 have occurred in this way. 



Aside from those dangers peculiar to fog there remains a number 

 of others. A continuance of cloudy weather or abnormal ocean 

 currents, or both, may throw the navigator out of his reckoning and 

 place him on a rocky shore a score of miles away from the safe route 

 he assumes himself to be following. 



Icebergs still remain a menace in spite of all the efforts which 

 have been made to guard against them. From time to time statements 

 have been made that apparatus has been devised which is capable of 

 locating their presence, but in every instance in which such apparatus 

 has been tested it has proved a failure. 



The history of systematic marine protection by means of light- 

 houses and beacons does not go back very far. It is true that there 

 were a few lighthouses, such as the Pharos of Alexandria, centuries 

 ago. but even in quite recent years a European Government received 

 a petition for compensation from the inhabitants of a seacoast district 

 on the ground that the erection of a lighthouse had deprived them 

 of one of their principal sources of income, to wit, luring vessels on 

 near-by shoals by means of false lights. 



The systematic employment of sound signals for marine protection 

 is of still more recent date and has never been carried out fully, in 

 spite of the fact that many of our greatest scientists, for example, 

 Tyndall and Eayleigh, have devoted special attention to this matter. 



One reason for this is that sound signals produced in air are very 

 erratic in" their range and intensity, so much so as to be on many 

 occasions absolutely misleading. This is due to the fact that when a 

 fog horn is blown the sound may be carried by the wind or may be 

 reflected or refracted by layers of air of different densities, with the 

 result that the sound may be audible many miles away, while there 

 may be a zone of complete silence extending from a few hundred 

 yards in front of the signal to a distance of 4 or 5 miles. 



As this phenomenon is by no means infrequent, the result has been 

 to discredit more or less this type of signal, and it will be evident 

 that the knowledge that a siren had been installed at a certain dan- 

 gerous point might prove a source of danger instead of a protection. 



As already stated, many eminent men have worked upon this 

 problem, but it was not until Arthur J. Mundy, of Boston, suggested 

 the use of water instead of air as the medium for transmitting signals 

 and proved its value by practical demonstration that any great 



