THE FISHERY. 497 



pool. Great have been the disputes as to whether the trout throw- 

 up these gravel hills with their heads or with their tails : the head of 

 the fish, however, shows much abrasion. 



It was years before I could see the act of spawning performed with 

 my own eyes, as apparently it is quickly done, and then possibly most 

 commonly at night. However, one gepial spring morning after a suc- 

 cession of cold nights, the fish gratified me by allowing me to see the 

 operation. I then saw distinctly the male fish chasing the female on the 

 spawn-bed, and the fish turning up the gravel at the tail of the spawn- 

 bed with their noses. When all the spawning is finished, some of the 

 male fish remain for two or more months at the tail of the spawn-bed, 

 and, as they chase away all strangers, I believe that they are watchmen, 

 playing the same part in the protection of the ova as the pugnacious 

 male sticklebacks undoubtedly do with respect to their nests. 



The artificial breeding of fish, or Pisciculture, has been followed 

 more or less in this country for many years, and Mr. S. Gurney 

 practised it on the Wandle at the Culvers. Mr. Sheppey was also 

 successful ; and the Wandle Fishing Association below the Culvers 

 now raise about 15,000 trout annually. 



The French Government, however, have of late years given an impetus 

 to the industry, and Professor Coste, who had charge of the subject, used 

 a gridiron of glass rods to support the ova in earthen vessels, over which 

 he allowed a current of water to trickle. I first learnt his process at the 

 College of France in 1859, and was highly delighted with the result. 

 The Professor was so obliging as to 

 give me specimens of young salmon 

 and trout, which I brought to England, 

 together with one of his boxes, in order 

 that I might make known his process 

 and use it in my garden, whence it 

 has spread over the country. 



The apparatus is simple. It con- f,c. .opB.-rish-.ray. 



sists of an earthenware oblong box (fig. 1098) about four inches deep, 

 six inches wide, and fourteen inches long, with a little spout. Inside 



K K 



