occasional cell of this type in normal blood is a 

 technic artifact, produced when the smear is made 

 by the mechanical stretching of the cell. On the 

 other hand, poikilocytosis does occur in chick- 

 ens; it is indicative of a disturbance in the blood, 

 and among the various shapes are many that have 

 a spindle form. Poikilocytosis obviously de- 

 velops within the bird and is not a technic artifact, 

 but this does not exclude the possibility that tech- 

 nic can play a role. Spindle cells have been seen 

 in the counting chamber of the hemocytometer 

 and, of course, under these conditions there 

 would be no stretching effect on the cells. 



Such spindle cells have been observed by 

 others, even in the embryo. Sugiyama (1926) 

 in his study says (p. 134), "It is noteworthy that 

 there are a few spindle-shaped red cells in the 

 blood of chicks, not only in embryonic life but 

 also after hatching. These red cells ordinarily 

 vary from medium size to exceedingly small, 

 sometimes with one end pointed and the other 

 rounded, sometimes with both ends pointed. 

 Such red cells have appeared by the time the em- 

 bryo has 22 to 29 somites, ahat is to say, from the 

 stage of early erythroblasts; at this stage they are 

 usually pointed at one end and rounded at the 

 other (figs. 17, 18)." His figure 17 is similar 

 to our figure 32 and his figure 18 is equivalent 

 to our figure 30. He goes on to say, "As an evi- 

 dence that such spindle-shaped red cells are by 

 no means to be considered as artificial products, 

 one finds them in the circulating blood widiin the 

 vessels of the area pellucida." 



Distortions of cells often produce rarefied 

 areas in the cytosome ; an example of this is shown 

 to a slight extent in figure 30 and more clearly 

 in figure 31. A variation in the production of 

 a spindle cell is shown in figure 32. One end 

 is round and the other drawn out into a long 

 tapering point. The fact that intermediate 

 stages between figures 31 and 32 can be found 

 leads to the suggestion that all these various dis- 

 tortions of cell shape have a common under- 

 lying cause. 



Distortion of cells may not necessarily pro- 

 duce pointed ends. There may be a slight ineak 

 in the side of the cell (fig. 34), a splitting apart 

 of the nucleus (fig. 33), a constriction of one 

 end of the cell (fig. 35), an elongation of the 

 cell (fig. 36), and the production of erythroplas- 

 tids of various sizes (figs. 41-43). 



The erythroplastid is probably produced by 



some process such as indicated in figures 35 and 

 38, and the anisocytosis sometimes observed 

 in cells could be accounted for by a diminu- 

 tion in size following the production of an eiyth- 

 roplastid. It is conceivable that a small one 

 (fig. 41) could be derived from a cell like figure 

 37, and a medium-sized erythroplastid (fig. 42) 

 from figure 40, and large ones (figs. 3, 2, and 

 43) from such a cell as figure 39. Primary 

 erythrocytes of embryonic blood break off anu- 

 cleated portions of cytosome more frequently 

 than do the definitive erythrocytes. 



Amitosis has already been mentioned in con- 

 nection widi the study made by Charipper and 

 Dawson ( 1928) on the blood of Necturus. They 

 include the elongations of cells and constrictions 

 of either nucleus or cytosome under their evi- 

 dence for amitosis, but, as already pointed out, 

 only rarely has there been any evidence of di- 

 vision of nucleus followed by division of cyto- 

 some. The opinion is held that they are ex- 

 tending tlie definition of amitosis too far when 

 they include the formation of erythroplastids. 

 They state, "Their fomiation may be considered 

 to be by an amitotic division of the erythrocytes 

 involving only the cytosome." Wilson (1925) 

 in his glossary defines amitosis as "mass-division 

 of the nucleus without the formation of spireme, 

 chromosomes or spindle-figure." Erythroplas- 

 tid formation is equivalent to the throwing off 

 of blebs of cytoplasm, which occurs so frequently 

 in lymphocytes, or to the pinching off of pseu- 

 dopodia. After discussing the evidence for and 

 against amitosis as a normal process of cell 

 nuiltiplication involving genetic continuity of 

 cells, Wilson (1925) states (p. 221), "It is clear, 

 therefore, that evidence of amitosis, unless based 

 on direct study of the living cell, must be re- 

 ceived with the greatest caution; . . ." This 

 point desei-ves reemphasis. The existence of 

 amitosis in avian blood can only be established 

 after careful in vivo studies of the type Speidel 

 (1932) has carried out on the tadpole tail and 

 Knisely, et al. (1947) on mammals. 



In birds, erythroplastids are relatively com- 

 mon and somewhere during the course of evolu- 

 tion from reptiles to the ancestor of the mammals 

 the process of erythroplastid formation became 

 fully estajjlished. It would be interesting to 

 know whether the survival value of such cell frag- 

 ments depended upon the more economical utili- 

 zation of space without the nucleus present or 



31 



