cells in other birds, he grouped all cells in which 

 the specific granules stained with eosin into two 

 types of heterophils: heterophils with ellipsoidal 

 rods and heterophils with bacillary rods. 



When the criteria set up in table 8 (p. 90) 

 are applied, as far as they may be applicable, 

 to Hewitt's colored figures of these two cell types, 

 the cells with the ellipsoidal rods become hetero- 

 phils and those with Ijacillary rods become eo- 

 sinophils. The close agreement between the dif- 

 ferential counts on ducks that he used and the 

 counts on the common mallard (p. 216) offer fur- 

 ther confimiatory evidence that his heterophils 

 with bacillary rods are eosinophils in spite of the 

 fact that they superficially resemble the hetero- 

 phils of other species of birds. 



The percentage values for leukocytes obtained 

 by Magath and Higgins (1934) for adult tame 

 mallard. Anas platyrhynchos L., differ somewhat 

 from tliose given in table 18. Like Hewitt, he 

 found that polymorphonuclears with granules 

 were more abundant than those with rods, namely, 

 24.3 and 2.1 percent, respectively. 



Figures 400^02 are from the mallai-d, figures 

 403 and 404 from the baldpate duck, and figure 

 405 from the ruddy duck. The three were se- 

 lected because they again illustrate the wide vari- 

 ation in the appearance of the eosinophil. Fig- 

 ures 403 and 404 are cells that were adjacent on 

 the same slide. The rods of the heterophil are 

 cigar shaped with a vacuole in the center of each 

 rod. The spheres of the eosinophil are very 

 large and most of them are homogeneous, but in 

 a few there are small granules on a reticulum; 

 the granules are arranged in a square. 



The type of eosinophil in figure 405 is a source 

 of confusion. It appears but little different from 

 the heterophil in figure 401. The specific bodies 

 of the eosinophil are present in the form of short 

 rods with granules. Sometimes these granules 

 are located at the end and sometimes in the middle 

 of each body. The eosinophil of the turkey (fig. 

 407) stains almost the same color as the hetero- 

 phil rods of the baldpate duck. A clear, light- 

 staining space is present in the center of some of 

 the spheres. 



Perhaps all these variations can be resolved 

 into some definite overall plan. Although much 

 study still remains, there seems to be some evi- 

 dence that the specific granule of the eosinophil 

 has two morphologic forms with transitional 

 stages between, not only in the same species but 



also in the same individual. In the chicken 

 there were small, fine granules on a reticulum 

 (fig. 179) and larger spheres, nearly homogene- 

 ous in structure (fig. 180). The condition in 

 figure 402 is equivalent to that in figure 179. 

 The important fact to note is that, at least in 

 chickens, the large type eosinophilic body is ac- 

 tually composed of four small granules arranged 

 in a square (fig. 180). Therefore, it is par- 

 ticularly significant that in the sphere located at 

 about 1 o'clock on figure 404 there should be 4 

 small, distinct granules in the form of a square 

 with homogeneous material around them. In 

 figure 405 they have taken another form, and here 

 the unit granules have moved apart in pairs so 

 that a rod is produced. 



These variations and their possible relation- 

 ships are presented in figure 411, in which the 

 progression is from the homogeneous sphere, 

 stage A, to the development of an enclosed tetrad 

 of granules, B, which in ducks may pull apart by 

 pairs to form pseudorod structures, B' and B". 

 The homogeneous sphere. A, may develop into a 

 square of four granules surrounded by much 

 matrix, B, or only a little, C. Step D has never 

 been observed but its existence, as a transitional 

 configuration between the scattered tetrads of C 

 and the reticulum with granules at the interstices 

 of E, is assumed. Any such scheme of pro- 

 gression should agree with the development of the 

 cell through myelogenic stages, maturity, and 

 aging. During myelopoiesis, the only expression 

 of the specific eosinophilic granule in the chicken 

 is in the form of homogeneous spheres (figs. 379- 

 381 ) . The same was true in a case of eosinophil 

 myelogenous leukemia. Of the immature stages 

 that were drawii from circulating blood (figs. 184 

 and 186), the first had homogeneous spheres and 

 the second a tetrad type of arrangement of small 

 granules. Since the second is a more differen- 

 tiated cell than the first, it might have been con- 

 cluded that the spheres were changed into gran- 

 ules, but when figure 186, a young cell with 

 granules, is compared with figure 180, a mature 

 cell with three nuclear lobes, it could he con- 

 cluded equally well that the process of change 

 was going in the opposite direction. 



Well executed color drawings of the blood 

 from an African vulture have been presented by 

 Neave (1906). He illustrates three morpho- 

 logic types of eosinophilic polymorphs, but does 

 not attempt to name them. 



208 



