38 



Appendices to Fourth Annual Beport 



use the term in tlie sense in which Kupffer used it, and can confirm his 

 observations on this point both from a study of the herring and also of 

 other forms. The nuclei in the parablast at this stage are not derived 

 from any pre-existing nucleus in the ordinary sense of the word. As 

 Kupffer remarks, they arise from dot-like beginnings and gradually 

 increase in size, and each is produced independently of its neighbour. 

 Whether the material of which the nuclei are formed is itself derived 

 in the first instance from the first segmentation nucleas of the ovum is 

 another question. Our knowledge of cells and cell divisions throughout 

 the animal and vegetable kingdom lead undoubtedly to the conclusion 

 that there is no such thing as true free cell formation, that every nucleus 

 must be derived from a pre-existing nucleus. There is no reason why 

 part of the original segmentation nucleus should not be contained within 

 the parablast, when this is separated from the archiblast by the formation 

 of the first equatorial furrow. As a matter of fact, I have not been able 

 to find such a nucleus, and have been led to suppose that its material is 

 generally distributed, and only again becomes aggregated in the form of 

 nuclei in the stage here described. After the formation of these parablast 

 nuclei the protoplasm around them becomes constricted off, so that 

 a number of cells are formed in situ. These then become pushed up 

 amongst the cells of the archiblast. At first they are easily distinguished 

 and stain differently, but after a time the difference is lost. The whole 

 of the parablast in this region is not used up in this way, but a part 

 remains which again increases in bulk, while the cells in the disc are 

 undergoing further division. A second batch of cells is added to the disc 

 in a similar manner, and we then have the stage shown in figure 8. 

 Cells next become distinguishable in the portion of the parablast which 

 has remained on the surface of the yolk. This collects around the 

 margin of the disc in the form of a thickened welt, which gradually thins 

 off over the yolk. The appearance in section is shown in figure 10. A 

 row of dot-like nuclei appears in the thickest portion, forming a complete 

 ring. The protoplasm around these nuclei becomes divided off, and a row 

 of cells is produced in the thickest portion of the peripheral parablast, as 

 shown in figure 9. The whole process is completely analogous with that 

 described at an earlier stage. Other rows of cells are added as develop- 

 ment goes on, until in the stage represented in figure 11 the peripheral 

 parablast has become cellular from the base of the disc to the equator of 

 the egg. Between the stages shown in figures 9 and 1 1 the disc begins 

 to extend around the yolk. This is accomplished by a thinning out of 

 what was originally the thickest portion of the disc. In figure 1 0 it will 

 be seen that the base of the disc is flat ; the process of extension can 

 easily be followed from a comparison of figures 11 to 14. As the disc 

 begins to spread over the yolk it no longer lies on the parablast in the 

 centre, and a space is thus formed between the two layers, which is the 

 segmentation cavity. This segmentation cavity separates the animal 

 from the vegetative pole. In figure 1 1 the disc has extended over an arc 

 of 160°, and in its centre has now a thickness of '3760 mm. In figure 12, 

 where the disc is fast approaching the equator of the egg (it covers an arc 

 of 176°), the diameter of the central portion is only -1880 mm., and the 

 periphery is almost as thick as the centre. In the next stage (figure 13) 

 the gradual thinning out of the central portion of the disc is still more 

 marked, while the periphery is now the thickest portion. In the next 

 stage it is seen that in oi3tical section the disc is thicker on one side than 

 on the other. This is represented in figure 14. It will be noticed on 

 comparing the figures, that in its extension the disc has gradually spread 

 over the parablast until in the stage shown in figure 14 the whole of the 



