BULLETIN OF THE UNITED STATES FISH COMMISSION. 197 



freely about with the sand particles or owing to their getting coated 

 over with the sand and mud their development was arrested. I have 

 not yet determined finally if the development is arrested when the eggs 

 are detached while development is proceeding, but this seems ex- 

 tremely probable. 



When at Ballantrae I noticed that the trammel-nets secured often more 

 males than females. Mr. Wilson, fishery officer at Girvan, informs me 

 that the ripest fish are caught in the trammel-nets, while most of the 

 unripe fish are obtained in the drift-nets, and that at the end of the 

 fishing season there are about three males taken for every two females, 

 indicating not necessarily that the males are more abundant than the 

 females, but rather that the males remain longer on the spawning- 

 ground ; and Mr. Wilson believes that herring prefer quiet water free 

 from strong currents when spawning, and that when the weather is fine 

 the herring remain long upon the bank and deposit their spawn lei- 

 surely, but when there are strong currents they either hurry the spawn- 

 ing process or disappear into deep water. 



As to artificial fertilization and hatching I found, after many experi- 

 ments a,t Ballantrae, that the best results were obtained when both the 

 male and female were held under water while the milt and ova escaped, 

 i. €., when the natural process of spawning is followed. 



^n ordinary wooden tub was obtained and filled with sea-water. 

 Into this a small quantity of milt was expressed, the male being held 

 completely under water while the milt escaped. A glass plate was then 

 held about 4 inches beneath the surface of the water, and, the fe- 

 male herring being held about 1 inch beneath the surface, by gentle 

 pressure the eggs readily escaped in the characteristic narrow beaded 

 ribbon, and, by moving the fish over the surface of the glass, either 

 a close or an open net-work could be formed. At first, where one 

 loop crossed another, the eggs were two or more layers thick, but, either 

 owing to the weight of the eggs or the gentle currents set up in the 

 water, before a few minutes had elapsed the eggs formed a single and 

 almost continuous layer, the net- work arrangement having disappeared. 

 The plate was then allowed to rest for two or three minutes at the 

 bottom of the tub, and a few short ribbons of milt were again intro- 

 duced. After moving the plate once or twice across the top of the tub 

 in order to wash off any scales that were adhering, it was placed either 

 in a hatching or a carrying box. Many thousands of ova treated in this 

 way contain extremely active embryos, which are expected to hatch on 

 March 22 or 23. 



(Professor Ewart exhibited a number of specimens showing herring 

 eggs attached to stones, sea- weeds, and sea-firs, and some of the herring 

 fry hatched on March 24 from the eggs artificially fertilized on March 8.) 



London, March 27, 1884. 



