October 28, 1898.] 



SCIENCE. 



bll 



countless numbers of dead fish while cross- 

 ing the northern edge of the Gulf Stream. 

 Investigation proved that these were tile- 

 fish, and that they appeared on the surface 

 of the water for an extent of 170 miles in 

 length and 25 miles in width. A conserva- 

 tive estimate, made by Captain J. W. 

 Collins, placed their number at upwards 

 of 1,438,720,000. Allowing ten pounds to 

 each fish, there would be 288 pounds of 

 fish for every man, woman and child then 

 in the United States. In September, Pro- 

 fessor Baird chartered the ' Josie Eeeves ' 

 and sent her to the tile-fish grounds, that 

 he might ascertain to what extent the spe- 

 cies had been depleted ; but the vessel re- 

 turned without having found a single indi- 

 vidual. 



In 1883 the ' Albatross ' made further 

 search, but without success. 



In 1884 the ' Albatross ' made a more 

 careful investigation, but again without 

 success. 



In 1885 the same vessel searched from 

 Newfoundland to the Gulf of Mexico with- 

 out discovering the least trace of the Tile- 

 fish, though Munida, a species of Crustacean 

 upon which the fish was known to have 

 fed, was found in abundance. 



In 1886, 1887, 1888, 1889, 1890 and 1891 

 nothing new was learned. 



In 1892 Commissioner McDonald fitted 

 out the ' Grampus,' and on August 5th 

 trawls were set on the old tile-fish ground. 

 No fish were taken. On the 6th the trawls 

 were set again, and one specimen weighing 

 seven pounds was brought to the surface. 

 This was the first specimen that had been 

 seen since the mortality of 1882, ten years 

 before. The ' Grampus ' continued her work, 

 and in about two weeks caught a second 

 specimen which weighed thirteen pounds. 

 On September 17th one specimen was 

 caught, and on September 18th three speci- 

 mens were taken. No more were caught 

 until October 8th, when two were found 



oif the Delaware coast. Thus, in 1892, a 

 search of two months yielded only eight 

 specimens. 



In 1893 the ' Grampus ' resumed the 

 search throughout the months of July, 

 August and September and caught scatter- 

 ing specimens. 



During 1894, 1895 and 1896 no additional 

 information relative to the fish was secured. 



On February 8, 1897, the Schooner 

 ' Mabel Kenniston,' of Gloucester, was over- 

 taken by a gale on George's Bank and blown 

 120 miles toward the southwest. After the 

 gale, trawls were set in sixty-five fathoms 

 of water, and thirty tile- fish were caught. 

 These weighed from six to fifteen pounds 

 each. They were landed at Gloucester on 

 February 16th. 



On August 12lh, of the present year, the 

 ' Grampus ' left Woods Holl with a small 

 party of scientific men, and sailed to a point 

 about seventy miles south of No Man's 

 Land. At the first set of the trawl, eight 

 beautiful tile-fish were taken. The boat, 

 insufBciently equipped with lines and bait, 

 at once returned to the ' Station.' New 

 trawls were purchased and on August 30th, 

 ice and bait having been taken on at New- 

 port, she again sailed south. The following 

 morning, when the boat was only sixty miles 

 from Block Island, the trawls were set. 

 The first haul yielded seven fish; the second, 

 forty-seven, and the third, nineteen. On 

 the following day seventy-eight fish were 

 taken, many of them of large size, and the 

 vessel, now bearing 1,000 pounds, headed 

 for Montauk Point, where the fish were 

 given to the soldiers at Camp Wikoff. 



When one considers that the trawls were 

 short, provided with only a few hooks and 

 tended by only one dory, it would seem 

 that the fish are sufficiently abundant for 

 an ordinarily equipped fishing-smack, with 

 its miles of trawls, to secure a full fare in a 

 very short time. 



The tile-fish, since the mortality of 1882, 



