BIOLOGICAL OBSERVATIONS ON THE HERRING. G33 



whether this was solely caused by the lower temperature; for it is prob- 

 able that other causes aided iu bringing- about this result, for example, 

 Iceeping the eggs for more than one and one-half months in small ves- 

 sels which could not be thoroughly cleaned; the impossibility of keeping 

 the temperature exactly at 0° (32° F.) ; and, finally, the change of water, 

 which coidd only be effected once a day. I do not maintain that it is 

 utterly impossible to produce perfectly healthy fish at a temperature of 

 Oo (320 Y.), for, iu repeating the experiment with better apparatus, some 

 of the mistakes of the first experiment might be rectified ; but a renewed 

 experiment would scarcely seem profitable, because it has been ascer- 

 tained as a fixed fact that at a temperature of -f 1^ C. (33.8^ F.) the 

 eggs of the herring develop in a perfectly normal manner, whilst repeated 

 experiments have shown that this is impossible at a temperature of 

 only — 0^.8 C. (30.56'^ F.). At this temperature the yolk becomes opaque, 

 expands, and finally bursts the shell of the egg. 



In the water of the Baltic, which is not very salty, the dividing line 

 lies between + 1^ (33.8^ F.) and — 0'^.8 C. (30.50^ F.); at any rate, very 

 near to zero. 



I have so far not been able to ascertain whether this condition would 

 remain the same in the water of the iSTorth Sea, which has a greater 

 degree of saltness, and whose freezing-point is lower. 



The fact that the spawn of the herring can stand such a low temper- 

 ature sufliciently explains why young herrings sometimes make their 

 appearance in the Schlei immediately after the breaking up of the ice. 

 They can, without any risk, lay their eggs even in very shallow water, 

 as no thick cover of ice, which alone might prove dangerous, forms in 

 spring. Even those eggs which have been laid earliest do not fully 

 develop until the water has become somewhat warmer. The autumn 

 herring never spawns in shallow water, but only where there is a cur- 

 rent. In the Western Baltic, therefore, the young herrings will scarcely 

 be destroyed by cold. On the otiier hand, it might be of interest to 

 investigate whether currents coming from the Polar Sea during the 

 spawning-season of the herring could strike the spawn in northern wat- 

 ers, for instance, on the coast of Norway. This would furnish an ans^fer 

 to the mysterious problem why the herrings leave certain coasts which 

 they had been iu the habit of visiting for many years. The surface 

 temperature cannot decide the question, but the temperature of the 

 water at the depth at which the eggs are found. 



In the above-mentioned report of the commission (p. 241), I have 

 mentioned the fact that iu the eggs of the autumn fish the yolk dimin- 

 ishes in size if the season of development is extended on account of the 

 cold. But as the autumn eggs used in former experiments differed in 

 size, and the young fish hatched from them in length, it was impossible 

 to decide whether those embryos whose time of development had been 

 prolonged by the cold had akeady increased in length whilst iu the egg. 

 As the spring eggs used in this year's exi^eriments were aU of the same 



