BULLETIN OF THE UNITED STATES FISH COMMISSION. 337 



served iu the cells of Salmo salar agree in all essential particulars with 

 those seen by Flemming in the cells of the skin and branchire of the 

 Salamander of Europe, and are of interest as confirming his observa- 

 tions and also as showing clearly for the first time that similar phe- 

 nomena occur iu the segmenting cells of a species of fish which has been 

 perhaps more thoroughly studied embryologically than any other. 



Fig. 1 represents a cell of the germinal disk of Salmo with its nucleus 

 in the quiescent or resting stage magnified about 800 times, which is 

 the amplification of most of the figures. In this condition the nucleus 

 is approximately globular and traversed by a very irregular and inter- 

 rupted network of aggregations, consisting of broken thread-like and 

 rounded granular bodies. This condition is the one which may be sup- 

 posed to exist during the long intervals of rest between the periods of 

 active segmentation which have been observed by different biologists 

 during the very early stages of development of the ova of various types. 

 This interval of rest may also include a portion of the stages of change 

 undergone by the nucleus preparatory to cleavage, and which will be 

 next described. 



Fig. 2 represents the nucleus of a cell from the germ disk of Sahno 

 which is elongating and undergoing a rearrangement of its contents 

 preparatory to division. With careful study it was found that nuclei 

 were present in different cells in this stage of metamorphosis. The 

 irregular broken granular threads and granules observed in Fig. 1 were 

 seen, with careful focusing, to have rearranged themselves in the form 

 of an intricate skein-like meshwork of threads, the direction of which 

 was generally more or less in conformity with the long axis of the nu- 

 cleus. This stage corresponds very closely with similar preparatory 

 stages fig-ured by Flemming and observed by him in Salamandra. 



Figs. 3, 4, and 7 represent other stages, also preparatory or progress- 

 ive. In Fig. 4 the nuclear matter exhibits a tendency to aggregate at 

 the opposite poles of the nucleus, while in Fig. 3 aggregations are taking 

 place around the equator of the nucleus as well as at the poles. The 

 equatorial ring of granular aggregations is viewed somewhat obliquely, 

 showing that they occupy a peripheral position against its wall. Fig. 7 

 represents a nucleus similar to the last, in which we may likewise note 

 the equatorial and polar accumulations of nuclear matter. Cleavage or 

 segmentation of the cell will take place across the plane marked by the 

 equatorial aggregations. 



Other phases in this strange cycle of phenomena are represented in 

 Figs. 8, 11, and 12. In Figs. 8 and 12 the nuclear equatorial plate 

 seems to be developed. In Fig. 11 the nucleus has apparently lost its 

 contour, but iu all of the last three the nuclear matter bas been aggre- 

 gated into definite stout threads, with enlargements at their ends in two 

 of the cases. 



In Figs. 5 and 6 the sharp contour of the nucleus has been finally 

 lost; short, thick, and highly refringent rods have been develoi)ed on 

 Bull. U. S. F. 0., 81 22 July 7, 1 S82. 



