448 Mr. E. E. Prince on the 



altered by an inherent power of movement which the consti- 

 tuent protoplasm possesses. To this movement are due the 

 creases and furrows continually diversifying the surface of the 

 blastomeres during cleavage, as well as the retrogressive 

 process by which blastomeres reunite occasionally after cleav- 

 age. Irregularities in segmentation are far from unfrequent*, 

 fourteen or eighteen cells being produced, and the outline of 

 the segmenting blastodisc is thus varied, though the circular 

 contour is always restored when the multicelled stage is reached. 

 The blastomeres thus do not always increase with that serial 

 regularity of geometrical progression which typical segmenta- 

 tion illustrates. Cleavage in a plane parallel to the upper 

 surface of the disc commences when the blastoderm is multi- 

 celled, {. e. consists of from fifty to eighty blastomeres, and 

 the form of each cell is altered by the increased pressure of 

 adjacent cells ; the original rounded or amorphous outline being 

 lost it becomes polygonal. The constituent cells simply con- 

 sist of naked protoplasm, and at the same stage vary very 

 much in dimensions, exhibiting a large, clear, more or less 

 central nucleus, which is not, however, always distinguishable ; 

 indeed there is evidence to show that periods when the 

 nuclei of the blastoderm and periblast are visible alternate with 

 periods when the nuclei are diaphanous. A similar rhythmical 

 alternation may be observed in the process of cleavage, a 

 discontinuity marked by alternations of activity and quies- 

 cence. 



Segmentation is not confined to the limits of the disc, but 

 at the margin the protoplasm of the periblast forms elevations 

 between which the lines of cleavage extend, and cells are thus 

 outlined in the investment of the yolk, which without doubt 

 must be added to the blastodermic mass. 



The Periblast. 



Cells formed, as just described, beyond the margin of the 

 disc contribute to the increase of the embryonic area ; but 

 such periblastic additions to the blastoderm apj)ear to be very 

 limited, and it is rather by an imperceptible process of intus- 

 susception that its increase must be accomplished and the 

 decrease of the vitellus accounted for. The lines of cleavage 

 cannot be traced far over the surface of the periblast ; they are 

 most distinct in proximity to the periphery of the disc, and 

 more remotely they pass insensibly away. The nuclei of 

 the periblast, which are more or less oval and well defined 



* At St. Andrews irregularity was most frequent in the ova of T, 

 (jurnardus ; but Ryder noted the same feature in the cod (U. S. Fish. 

 Comm. Rep. 18b2,'pp. 486-7, pi. ii. %. 12). 



