[3] OPERATIONS AT THE NORTHVILLE HATCHERY. 1039 



Kurrouuded by au impervious mass of cement and gravel. Until this 

 is done it will not be possible to command the entire yield of the sup- 

 ])lying springs, nor to dictate through what channels the water shall 

 be discharged. 



The cooling or intermediate reservoir between the spring pond and 

 hatchery had been leaking quite too freely to be compatible with safety, 

 so that repairs were considered essential. We therefore girted it with 

 a 10-inch band or rim of 2-inch planking, the water level touching the 

 middle of the rim, while the planks themselves are tirmly held against 

 the outer wall by spikes driven to stakes set in front. Then, a double 

 «)at of cement, lapping on the edge of this rim and covering the entire 

 interior surface of the reservoir, was spread, thus eflectually closing all 

 possible chances for leakage. The test of eight months' use of this re- 

 <epta(^le has shown it to be absolutely Avatertight and i)erfectly safe. 



NOTES AND TABLES IX REFERENCE TO COLLEOTINO THE SUPPLY OF 



WHITEFISH EGMJS. 



Most of the eggs laid in were secured at the "Bass'' islands of Lake 

 Erie, which are, on the whole, quite as reliable as any locality for this 

 work. Certain other points in Lake Erie, as well as in Lakes Huron 

 and Michigan, may show heavier catches of fish, but the}' are, so far as 

 1 have been able to ascertain, less prolific of ripe fish, in proportion to 

 the number caught. At the islands, too, as well as at all other points, 

 the yield of eggs from the various fisheries is quite disproportionate to 

 the catch of adult fish. The fact that wbitefish are caught in any given 

 locality during their nominal spawning period does not necessarily sig- 

 nify that ripe fish will be found at such places, for the devices for their 

 capture — the stationary tiap net or portable gill-net — may not be set 

 on or near those grounds naturally selected by the fish for the deposi- 

 tion of spawn, but at pointvS in the paths or runways leading to and 

 quite remote from the objective point of the fish in their migrations 

 from the feeding to the spawning grounds. Ftom these nets ripe fish 

 are found, if at all, with the exception of an occasional straggler, in the 

 later runs at the last of the season. 8uch fisheries, although quite 

 profitable for the fishermen, are generally unreliable for the collection 

 of spawn, especially when adverse weather compels a suspension of 

 work before tlie last migrations occur. 



There are certain spawning grounds in the vicinity of the islands that 

 can invariably be depended on. These are well known, and have be- 

 come favorites with the spawn-gatherers, not only Ix^cause of their reli- 

 ability and certainty of being visited by schools of ripe tish, but ripe 

 tish usually appear several days earlier than at other points, some of 

 which furnish heavier catches. Indeed, the privilege of collecting eggs 

 from the n<?ts set on these fruitful grounds is so much sought after by 

 the representatives of various fish commissions that, naturally enough, 

 considerable rivalry for the control thereof is developed. Naturally 



