ATKINS THE SALMON AND ITS ARTIFICIAL CULTURE. 249 



one beiug the way by which it eutered, which it woukl not be likely 

 to fiud ; the other through the gate into the brook, to which it would 

 hav^e easy access. This structure was completed on the 9th of October, 

 and the large number gathered within it before the close of the month 

 attested its usefulness. 



Second. A seine was prepared, 18 feet in depth and about 1,000 feet long, 

 which was of suflicient size to span the pond at its widest part ; and this 

 was twice drawu the whole length of the pond toward the dam to drive 

 the salmon inside the hedge. Tiiis appears to have done some service, 

 although the character of the bottom, which was soft and studded with 

 an abundance of snags, prevented its being as effective as it would have 

 been on a smooth, hard bottom. 



Third. On the two brooks that are tributary to the pond barriers 

 were constructed to prevent the ascent of the salmon, and in the larger 

 one at Redman's Bridge* a ])onnd was made to entrap any that might 

 try to pass that point. These barriers and pounds might have been 

 successful had it not been for a heavy fall of rain, almost beyond precedent, 

 that raised the water quite over their tops and let a large number of 

 salmon pass up the larger brook. 



Early in October the salmon showed an increased disposition to rove. 

 There had been sufficient rains to flow the meadows adjoining the pond, 

 and several salmon were observed there about the eighth day in very 

 shoal water. Immediately after the hedge near the dam was completed, 

 the fish began to gather in it, exhibiting great activity, running and 

 leaping. Their manner of leaping, however, was different from that 

 observed in summer. Then the leap was almost vertical in the air ; now 

 it was generally more oblique, describing the arc of a large circle. I 

 have seen a salmon in October leap through the air a horizontal distance 

 that I judge to be ten feet, without rising more than three feet from the 

 surface of the water. Such movements suggested that they were either 

 playing or fighting with each other, and that the leaps were accidentally 

 made by salmon that were deluding the pursuit of others. 



On the 24th day of October the grating, which up to this time had 

 been kept before the gate, was removed, and a free but small passage 

 was left open for the salmon. One very small male had already got 

 through the grating into the brook, and his milt was found to be mature. 

 Up to tJae evening of the 26th no more came down. That night it rained 

 so heavily as to raise the pond. . The next day, the 27th, a number of 

 salmon were seen to pass through the gate into the brook. It was gen- 

 erally observed that they descended tail first. That was the case in 

 every instance that I observed. The gateway was only eight inches 

 wide, and some of the larger fish, turning partially on their sides as they 

 dropped into it, could hardly get through. It was afterward enUirged. 

 On the 28th the taking of spawn began. The first eight females ex- 



* See map of Spofford's Brook. 



