790 REPORT OF COMMISSIOJ^ER OF FISH AHD FISHERIES. 



trays were employed about 12 inches square, nested in frames carrying 

 20 trays per frame — the identical apparatus used at this establishment 

 in 1875 and yearly since. The water was taken from the open stream 

 through a covered plank conduit, with the expectation that very pure 

 water would thus be secured. It was afterwards found that the little 

 brook tliat flows from the old hatching-house through a swampy piece 

 of laud discharged its waters into the stream above the new house in 

 such a way that, instead of mingling at once with the water from the 

 lake, they crept down along the shore almost by themselves, as far as 

 the hatching-house. It thus came about that whenever the brook was 

 in flood its turbid waters crowded the pure water of the lake away from 

 the conduit, and took entire possession of the hatching-troughs, making a 

 very dirty jjiece of work of it. It is not known that any harm resulted 

 beyond the extra work involved in cleaning up and the unpleasant ap- 

 pearance of the fixtures. But steps have been taken to avoid such an 

 occurrence in future by continuing the conduit out under the water of 

 the stream far enough from shore to avoid receiving any part of the 

 brook water. It will be seen that this new hatching-house, though of 

 the greatest service as supplementary to the old one, could not wholly 

 take the place of the latter, which alone afforded facilities for hatching 

 out the reserve for the stream. I therefore turned my attention to the im- 

 provement of the old house. In the first place, it appeared advisable tO' 

 secure, if possible, better aeration of the water ; for this end the situa- 

 tion was a very unfavorable one. The spring issued from a gravelly bank, 

 at an elevation scarcely above the level of a large swamp, through which 

 the overflow oozed away. We had the year before cut a wide and deep- 

 ditch, nearly half a mile long, for the outflow, so that there was no 

 longer danger of the house being flooded by fjseshets, but this did not 

 enable us to lower the Roughs from their original elevation. We could 

 not curb the spring and thus raise a head, because of the danger that 

 the water would then find a new outlet through the loose gravel and be 

 lost to us altogether. The available head was thus scarcely a foot. The 

 best that could be done was to construct above the hatching-house a 

 narrow, circuitous tlrain or canal, about 10 feet wide and nearly 70 feet 

 long, through which the water should flow with a surface air-exposure 

 of about 140 square feet, before entei-ing the hatching-house ; to have aU 

 the overflows and passages, from canal to feeder and from feeder to 

 hatching-trough, at the surface rather than submerged ; and to intro- 

 duce in all the troughs occasional dams which should bring all the water 

 repeatedly to the surface and expose it to the air in wide and shallow 

 currents. 



Careful search also revealed a very considerable leak around one end 

 of the hatching-house dam. This was finally, though not without some 

 difficulty, completely stopped. No other change of importance was made 

 in the general hatching arrangements. 



The summer, and more especially the early autumn, were rather dry 



