160 FISHING GROUNDS OF NORTH AMERICA. 



bordering the coast, cousiclercd also in their relations to the temperature of the air. The outside 

 light stations selected for taking the observatious are twenty-four in number, and represent the 

 entire eastern coast of the United States from Southern Florida to Eastern Maine. As enumer- 

 ated and described below, it will be noticed that some are light-ships anchored ofl" the land, while 

 others are light-houses, situated on islands or on the main coast. Those of the first class are 

 infinitely better located for temperature purposes than either of the others. At most of the 

 stations observations were begun as early as 1878, but prior to 1881 so many breaks occurred iu 

 the records that it has been found inexpedient to make use of any of the data relating to the first 

 three years. This rei^ort, therefore, covers a period of only five years, from 1881 to 1885, inclu- 

 sive. Although the depths at the different stations vary greatly, ranging from a few feet to 18 

 fathoms, it was thought advisable to have the observations cover the bottom as well as the sur- 

 face waters, but the former having been entirely neglected at nearly every station, no reference is 

 made to bottom temperatures on the following pages. 



The record blanks supplied to the light-house stations call for the following observations: 

 Depth of water where the observations are taken, at mean low tide; time of observatious, which 

 are to be made twice each day at the first high water and first low water after 7 a. m.; tempera- 

 ture of the water at the surface and at the bottom, and by exposing the thermometer in the open 

 air; direction and force of the wind, and state of the sky at the time of taking the temperature 

 observatious ; occurrence and duration of rain or snow ; occurrence and movements of any kinds of 

 fish, singly or in schools, especially menhaden, herring, cod, mackerel, swordfish, horse-mackerel, 

 bluefish, &c., and also of seals or whales. 



As the observations are taken at the first high water and first low water after 7 a. m., they 

 are not made at the same time every day, but generally fall within the twelve hours following 7 

 a. m., or between 7 a. m. and 7 p. m. Should the time for taking the first observation come 

 immediately after 7 a. m., the second observation would be made soon after noon. As, iu some 

 places, there is considerable difl'erence between the temperature of the water at high tide and low 

 tide, this method of regulating the hours of observation appeared to aiibrd the most satisfactory 

 means of equalizing results. As elsewhere explained, the temperature observations were limited 

 at most stations to the surface water and the air. Observations respecting the direction and 

 force of the wind, the state of the sky and the occurrence of rain were generally well kept, but 

 the opportunities for observing the movements of fishes were not equally good at all stations, nor 

 was it to be expected that a constant outlook would be maintained by volunteer observers already 

 burdened with other and more important duties. Nevertheless many interesting and valuable 

 records were made in that line, although nothing of importance was learned respecting the I'egu- 

 lar migratory movements. 



The thermometers employed were of two kinds, both of which were especially designed for 

 taking water temperatures not only at the surface, but also in slight depths. During the first 

 few years the stations were supplied with the excellent pattern made for the Signal Service, and 

 kindly lent for the purpose. These thermometers are inclosed in a stout bronze case, with suit- 

 able openings for the entrance of water, and every instrument was carefully comiiared with a 

 standard before being sent out. From time to time the Signal Service thermometers have been 

 replaced by a new and equally reliable pattern, made by Charles Wilder, of Peterboro', N. H., I'or 

 this special kind of work, and at present all the light-house observations are taken with these 

 instruments. The tube is protected in a cylindrical copper case, somewhat similar in construction 

 to that of the Signal Service, but of larger size. All instruments are compared and none having 

 a large error are made use of. Occasionally, at some of the stations, when the regulation ther- 



