506 MARGARET MORRIS 



cases that a definite number of chromosomes which he was often 

 able to identify as paternal in origin were eliminated in the early 

 cleavages. When this was the case, the larvae showed only 

 maternal characters. When the male chromatin was retained 

 the larvae were intermediate in character, except in the cases of 

 Echinus 9 X Antedon d' and Strong. 9 X Antedon d" where 

 there is no elimination of chromatin. In this case, nevertheless, 

 purely maternal hybrids result. Tennent ('12) found much the 

 same condition in hybrids between Hipponoe and Toxopneustes. 

 In Tox. 9 X Hip. d there is no elimination of chromatin, and the 

 male dominates; while in the Hip. 9 X Tox. d cross, in which the 

 maternal characters dominate there is elimination of a certain 

 amount of chromatin which, although it cannot be traced 

 definitely to either parent, is presumably of paternal origin. In 

 crosses between Arbacia and Toxopneustes ('12 b) he finds that 

 some chromatin of female origin may also be eliminated. 



In the hybrids with which the present paper is concerned, we 

 find a condition differing somewhat from any of these, though 

 naturally most like that described by Moenkhaus for the Fundu- 

 lus X Menidia cross. As has been said, it is necessary to draw 

 all conclusions from the Fundulus eggs as the material obtained 

 from the reciprocal cross was unsuitable for study. The first 

 stage of interest in the hybridized eggs of Fundulus is that 

 represented in figures 1, 4 and 5, where the germ-nuclei are closely 

 apposed, and the aster has divided. In figures 4 and 5 the asters 

 he at opposite poles of the pair of nuclei, while in figure 1 they 

 have not yet taken that position. Tt will be noticed also that 

 in one case the nuclei lie side by side, and in the other end to 

 end. In many cases one can see a difference between the two 

 nuclei in the density of the reticulum, but this is not universal, 

 and it is impossible to say with any degree of certainty which 

 is the male and which is the female. A study of the normal 

 Fundulus egg shows that this failure of the germ-nuclei to unite 

 is peculiar to the hybrids, and that the stage represented in fig- 

 ures 1, 4 and 5 is in fact that which immediately precedes the 

 formation of the first cleavage spindle. The condition of the 

 normally fertilized eggs is shown in figure 3. As long as the two 



