cause such cells were choseu in order to make 

 more convincing the evidence that these are 

 tlirombocytes and not lymphocytes. Lympho- 

 cvtes, monocytes, and granulocytes are not nor- 

 xiialiy present in circulating hlood of the embryo, 

 although cells belonging to these types may be 

 ^een occasionally, and they will be considered 

 more fully when the cells found within the hema- 

 topoietic organs are described. 



It has been indicated already that mature 

 stages from later generations of thrombocytes, 

 like later generations of erythrocytes, approach 

 a higher level of terminal differentiation than did 

 earlier generations and thus during the second 

 week of embryonic development, all cytologic 

 features that characterize the definitive cell can 

 be seen but not exactly as they will appear in the 

 circulating blood of the adult bird. Specific 

 granules of the type shown in figure 296 are 

 atypical in that they are larger than normal and 

 in that each lies within a vacuole. This vacuolar 

 effect may possibly be due to technic since it is 

 found also in mature thrombocytes from the cir- 

 culating blood of the adult fowl, when the blood 

 smear has been fixed in Petrunkevitch No. 2 and 

 stained with May-Griinwald Giemsa (fig. 202). 

 The general appearance of the cell (fig. 296), 

 the lightly stained vacuolated cytoplasm, the 

 condensation of the nucleus, and the acidophilic 

 affinity of the cell margin, all establish this par- 

 ticular cell as an embryo thrombocyte. Typi- 

 cally tlie specific granules are distributed among 

 the strands of the cytoplasmic reticulum (figs. 

 297-299, 302, 303, and 305). 



The last-mentioned point, the affinity of the cell 

 perimeter for eosin, is especially significant. 

 This reaction by the cell margin is often seen in 

 embryo thrombocytes of this age. The same re- 

 action occurs in thrombocytes of adult birds (fig. 

 85), but after hatching, the disintegration proc- 

 ess is less frequently stopped at the exact moment 

 when this initial stage in cytoplasmic breakdown 

 would be revealed. 



Other examples of crumpling and acidophilic 

 staining of the cell margin, an initial reaction in 

 the disintegration process, are shown in figures 

 297 to 299. There is merely a single indenta- 

 tion of the cell wall in the first two of these fig- 

 ures, but in the third the entire margin is scal- 

 loped. There is a question of whether the shift 

 from a basophilic to an acidophilic affinity (1) 

 was a result of cell disintegration, (2) preceded 



it, or (3) was under the control of a different set 

 of factors with little or no causal relationship to 

 cellular breakdown. For the present, only ex- 

 amples can be mentioned, and these do not al- 

 ways support the same hypothesis. Figures 302 

 and 303 show a definite acidophilic margin and 

 slight diffusion of the reaction into cell proto- 

 plasm; yet there is no indication of cell rupture. 

 On the other hand, figure 306 shows cytoplasmic 

 disintegration well started, yet basophilic affin- 

 ities have been retained, and figure 295 shows 

 both rupture and acidophilic staining. These 

 examples come from different slides. The cyto- 

 plasm of thrombocytes under the best conditions 

 stains so delicately that only with optimum illu- 

 mination correctly aligned can these differences 

 in color and structure be detected. 



The cell represented by figure 302 was selected 

 to show that an intact, almost definitive, throm- 

 bocyte can be found as early as 9 days 15 hours. 

 This cell was not seen until after a number of 

 slides had been made. This degree of differenti- 

 ation is not typical for this age. Usually small 

 thrombocytes of mid-embryonic life look like 

 figures 300, 301, and 306, or those of figures 

 228 and 229. Cells (figs. 300 and 306) were 

 selected for drawing in which specific granules 

 were still visible; when these granules are not 

 visible, disintegrating thrombocytes so closely 

 resemble small lymphocytes that have dis- 

 charged part of their c^tosome, that the identifi- 

 cation might be questioned. Some rounded 

 cells (fig. 304) in which specific granules are 

 absent but in which disintegration has not begun 

 can definitely be identified by the lightly stained 

 vacuolated cytosome. If the technic methods 

 could be improved, probably most of the embryo 

 thrombocytes after 10 days of incubation would 

 look like figure 302. 



When a cell reaches a stage of degeneration 

 such as shown in figure 301 and a specific gran- 

 ule is not visible, one can only guess at its iden- 

 tity: if it is part of a cluster (fig. 229, 19) it is 

 probably a thrombocyte, but if it is an isolated 

 cell, it may be a lymphocyte. Generally- how- 

 ever, the nucleus of even the small lymphocyte 

 is larger than that of the thrombocyte. 



Even at the end of the second week of incuba- 

 tion, thrombocytes in general have not taken on 

 the definitive form and many, even with specific 

 granules, still have an appearance of immaturity 

 (fig. 305) . This may be true until nearly hatch- 



132 



