larrabee. — the optic ciiiasma of teleosts. 229 



The Effect of Gravity upon the Crossing of the 

 Optic Nerves. 



0. Schultze ('95) in some experiments with frogs' eggs, found by 

 compressing the eggs between glass plates so as to force the blasto- 

 derm to remain at some point other than its normal one, that he 

 produced abnormal specimens. The degree of abnormality varied 

 according to the angle made by the axis of the normal egg with the 

 attraction of gravity. I have carried on similar experiments with 

 eggs of the brook trout, but with no results so far as abnormalities 

 were concerned. The compressed eggs died within a week or ten 

 days. But the experiments showed conclusively that gravity could 

 have no effect on the development of the nerves, and therefore could 

 not cause the dimorphism ; for in all cases the blastoderm finally came 

 to the top of the egg, in spite of the pressure. If the blastoderm 

 remained at the top under such conditions, it certainly would do the 

 same under normal conditions. In this position the effect of gravity 

 on the two sets of developing nerve fibres would be the same, and 

 consequently gravity could not be a factor in the question under 

 consideration. 



The Laws of Chance applied to the Results. 



The ■svriter has now shown that the dimorphism under consideration 

 is not hereditary, is not due to an earlier development of one of the 

 optic nerves, and is not caused by the action of gravity. It seems to 

 be begging the question to say that it is a matter of chance whether 

 one nerve is uppermost or not ; for this means that the real cause of 

 the dimorphic condition is unknown and, whatever it may be, it was 

 not affected by the conditions of the experiments performed. Admit- 

 ting that a " chance result " has this significance, the writer wishes to 

 show that the results obtained from the dissections are really chance 

 results. If so, the chances would be equal, in the normal trout, that 

 either the right or the left nerve should be uppermost. In other 

 words, the so-called rights and lefts should occur in equal numbers. 

 Referring to the results obtained from the first and second generations 

 of the trout, and to the control lot, also to the results obtained firom 

 the cod, it will be seen that the observed results closely approximate 

 such a relation. Of Salvelinus, 4950 specimens in all were dissected. 

 Of these, 2749 had the right nerve dorsal, and 2201 the left nerve dorsal. 

 Of Gadus, 1132 specimens were dissected; 580 of these had the right 

 nerve dorsal, and 552 the left nerve. In both cases, but especially in 

 the first, there is an excess of rights. In view of the results obtained 



