246 TEXT-BOOK OF PHYSIOLOGY 



number of red corpuscles, 5,000,000 per cubic millimeter, is also considered 

 as 100 per cent. Under such conditions the corpuscles have a normal color 

 known as the color index. This is expressed by a fraction of which the 

 percentage of hemoglobin is the numerator and the percentage of corpuscles 

 the denominator. The normal color index is therefore i or unity. In 

 some pathologic states the hemoglobin alone diminishes, the number of the 

 corpuscles remaining the same; in this instance the color index is less than 

 unity, e.g., if the hemoglobin be reduced to 80 per cent, as determined by the 

 method to be described, then the color index will be T ^ F = o.8 which indicates 

 that each corpuscle retains but eight-tenths of the normal amount of hemo- 

 globin, or stated in the reverse way, each corpuscle has lost two-tenths of the 

 normal amount of its hemoglobin. In other pathologic states there is both 

 a diminution in the percentage of the hemoglobin and in the percentage of 

 the corpuscles and the diminution may be equal or unequal in degree. If 

 the diminution' be equal the color index is unity; if it be unequal the color 

 index is less or greater than unity; e.g., if the percentage of hemoglobin be 

 but 60 and the percentage of red corpuscles, as determined by the method 

 of counting be but 80 (4,000,000 per cubic millimeter) then the color index 

 is 6-0 = 0.7 5 which indicates that each corpuscle retains but three-fourths 

 of the normal amount of hemoglobin; if on the contrary the percentage of 

 hemoglobin be but 60, and the percentage of red corpuscles be but 50 then 

 the color index is 1.2 which indicates that each corpuscle contains a larger 

 percentage of hemoglobin than normally. This condition is sometimes 

 observed in pernicious anemia. 



For the determination of these variations in the hemoglobin for clinical 

 purposes two chromometric methods are at present largely employed, that 

 of Gowers and v. Fleischl. All chromometric methods are based on the 

 principle that if two equally thick and equally well-illuminated solutions 

 present the same intensity of color, their richness in coloring-matter is the 

 same. There are two methods by which this can be done: (i) By diluting 

 the blood to be examined with water until the shade of color corresponds 

 to that of a solution of hemoglobin of known strength (Gowers). (2) 

 Diluting a given quantity of blood with a given quantity of wa ter and then 

 finding an identical color which represents a previously determined quantity 

 of hemoglobin (v. Fleischl). 



Gowers' hemoglobinometer consists of two glass tubes of exactly the 

 same size and similar to those shown in Fig. 99. One, A, contains glycerin 

 jelly colored with picro-carmine the shade of which corresponds to that of 

 normal blood diluted 100 times (20 c.mm. in 2000 c.mm. of water), repre- 

 senting a i per cent, solution. The other tube, B, is ascendingly graduated 

 with 120 divisions, each one of which corresponds to 20 c.mm. With a 

 graduated pipette, D, 20 cubic millimeters of blood are accurately measured 

 and dropped into the bottom of the tube B, in which a few drops of distilled 

 water have been placed so as to prevent coagulation. Water is then added 

 drop by drop until the color of the diluted blood is exactly that of the stand- 

 ard. ^ The division of the scale reached by the dilution will represent the 

 relative percentage of hemoglobin. If this tint is not obtained until the 

 dilution reaches 100 divisions, the quantity of hemoglobin is normal. If 

 more water must be added, it is in excess; if less, it is diminished. If, for 

 example, the 20 cubic millimeters of blood from an anemic patient gave the 



