4(j8 PROCEEDINGS OF THE NATIONAL MUSEUM. vol. xxix. 



sea water. From a third lot of eggs the water was carefully drained, 

 and over them was poured sperm from testes which had been torn up 

 in a perfectly drj^ dish. These were allowed to stand for a few min- 

 utes, and were then placed in clean, running sea water. The females 

 were certainly ripe for spawning, and the males were well grown and 

 had not recently borne eggs, so they were presumabl}- fertile. A con- 

 trol experiment was made by putting a batch of this last lot in running 

 sea water without the addition of sperms. In all cases the results 

 were the same. At the end of one and one-half hours protoplasm 

 could be seen collecting at the upper pole. After two to three hours 

 it was noticed that the eggs had flattened slightly at the animal pole 

 and that there was being formed a pretty clearl}^ defined round ger- 

 minal disk, resting on a \sijev of orange-red oil drops. At the age of 

 four to six hours the germinal disk was at its prime, but neither then 

 nor at an}^ subsequent time was there any trace of segmentation. 

 From this time on the germinal disk gradually lost its sharp outlines, 

 flattened down, and went to pieces. In one lot of eggs at the age of 

 twenty-six hours the germinal disk had gone bad; in another after 

 twentj^-five hours it was no longer round, and its edges were irregular 

 and fragmentary; in a third lot less than 10 per cent of the eggs were 

 alive after twenty-three and one-half hours. 



These eggs were all alike save that in one lot some, when taken 

 from the ovary, showed a very faint aggregation of protoplasm at the 

 germinal pole, while in another lot the eggs were of unequal size. 

 This latter condition is, however, b}^ no means an uncommon occur- 

 rence. Such differences are met with repeatedly in my preserved 

 material, where eggs one- half to two-thirds the size of the normal 

 ones are found. Save that the blastoderms are somewhat smaller, 

 there is nothing unusual about the development of these small eggs. 

 In this connection Brook (1887) says that the eggs of the herring 

 vary in size in the same tish or in fishes of different localities, but 

 thinks that this in no wise affects their development. 



From ni}' experiments it seems pretty clear that artificial fertiliza- 

 tion is not possible in the pipefish, thus confirming the aj/no/'l opin- 

 ion that this would not take place in fishes provided with such extraor- 

 dinary apparatuses for the deposition and impregnation of the eggs, 

 without their ever coming in contact with the water. Since the eggs 

 will live for some twent}' hours in sea water, it must be the sperma- 

 tozoa w^hich are disastrously affected by it. It has long l)een known 

 that the sperms of both salt- and fresh- water fishes lose their vitality if 

 left in the water any time and can not impregnate eggs. Quatrefages 

 first ascertained this for the pike and other fresh-water fishes. Hoff- 

 mann (1881) sa^'s that the sperms of !Sc()r])cV)i(( die quickh^ in salt-water. 

 Reighard (1893) found that the sperms of the wall-eyed pike die after 

 one minute in the water. 



