with avian blood that it is impractical. In mam- 

 malian blood only the leukocytes have nuclei, 

 and so any smudged nucleus must at least be a 

 leukocyte. But in avian blood where the ery- 

 throcytes and thrombocytes are nucleated also, 

 the possibility of error in identification of a 

 smudged nucleus becomes so great that any pos- 

 sible value of its addition to a count is nullified 



by the increased source of error ; and if some can- 

 not be properly classified, those that can be iden- 

 tified, such as the three granulocytes, should not 

 be included either. On the other hand, if in a 

 particular bird the smudging affects only one of 

 the types of white cells, some account must be 

 taken of the fact, or the difterential values will be 

 biased. 



Granular Leukocytes 



There are three granular leukocytes in birds as 

 in mammals — heterophils, eosinophils, and baso- 

 phils. Eosinophils and basophils received their 

 names because the cells contain granules that 

 have an affinity for eosin and for basic dyes. 

 The term "heterophil," which was suggested by 

 Kyes (1929), applies to the third granulocyte 

 cell type, in which the specific inclusions of ho- 

 mologous cells among the various classes of ver- 

 tebrates show great diversity in reaction to stains. 

 The heterophil of birds and reptiles is the 

 equivalent of the neutrophil in man. The term 

 "neutrophil" is based on the staining reactions; 

 the temi "heterophil" is not. 



Ryerson (1943), from his comparative 

 studies, suggests that the morphology of hetero- 

 phils and eosinophils has been influenced by two 

 lines of evolution through the vertebrate classes 

 (p. 44)," . . . one line contains the selachians, 

 reptiles, and birds; the other contains the cyclo- 

 stomes, teleosts, amphibians, and mammals." 



HETEROPHILS 



Normal mature heterophils (figs. 154— 

 167) 



Heterophils as usually seen in the circulating 

 blood show a low variability in size. Some in- 

 dication of range may be obtained by comparing 

 the various drawings and, particularly, the graph 

 (fig. 197). Kennedy and Climenko (1928) 

 gave a range of 4.2 to 9.0/u with an average of 

 6.35m, which is nearly 2.3m less than the average 

 of 8.7m from these data. Under stimulation 

 such as replacement after irradiation, a shower 



of smaller cells may be liberated. Perhaps this 

 reflects a factor of cytoplasmic growth, independ- 

 ent of seeming maturity that is indicated by nu- 

 clear lobulation, in that smears made from bone 

 marrow show many more small heterophils than 

 are found in circulating blood, and yet these cells 

 show multilobed nuclei and in this respect are 

 considered to be mature. This fluctuation in size 

 of heterophils, and also of eosinophils, is a factor 

 that may account for differences in size reported 

 by different investigators for these cells. If one 

 emphasized the size of heterophils as some have 

 emphasized the size of lymphocytes, it might be 

 that cell size would prove to be a criterion as use- 

 ful as number of nuclear lobes in indicating the 

 condition of health of the l)ird. 



From these figures of heterophils it can be 

 seen that the dominant shape is a circle, which is 

 sui-prising in view of the fact that in life heter- 

 ophils are actively amoeboid. 



In a strict sense practically none of tlie heter- 

 ophils from circulating blood should be regarded 

 as representing their true structure and appear- 

 ance. Nearly all these cells reveal an artifact 

 involving the nucleus, and many reveal a second 

 artifact that involves the rods. The artifact 

 in the nucleus is so constant with Wright's stain 

 that it is of considerable value in identifying the 

 cell type, but the one in the rods is an ever-present 

 cause of confusion between heterophils and 

 eosinophils. 



The rods, which are the specific inclusions for 

 heterophils, have typically the appearance shown 

 in figures 154 and 155. They are long fusiform 

 bodies, they are pointed at each end, and they 

 take the eosin stain brilliantly. Both Dant- 

 schakoff (1909a and b and 1916b) and Hamre 



73 



