THE JOURNAL OF THARM A.COLOGY. T25 



Therefore 48$ Fleischel would have color-index 2 — a case of per- 

 nicious anemia. 



Why color-index is 2 will he explained later. 



We often read and hear of percentage of hemoglobin in a case without 

 reference to the number of corpuscles. You can readily see that the 

 former, without reference to the latter, gives no information regarding the 

 reduction of hemoglobin. 



We have another method of estimating the amount of hemoglobin 

 which, owing to its simplicity, is worthy of mention. It is Hammerschlag's 

 indirect method of taking the specific gravity of the blood. With a me- 

 chanical mixture of two liquids, such that a drop of blood in it neither 

 sinks nor swims, we have a mixture of about the same specific gravity as 

 of the blood. So taking the specific gravity of the mixture we have, for 

 clinical purposes, the specific gravity of the blood. 



Technique. — It is very simple. We prick the finger and allow a large 

 drop of blood to drop into a mixture of chloroform three parts, benzol one 

 part (specific gravity 1.055=that of normal blood). If the drop sinks add 

 chloroform, if it swims add benzol, and so by adding one or the other, we 

 bring the drop in the middle, where it remains. Now take the specific 

 gravity of this mixture, which is approximately the specific gravity of the 

 blood. The solids of the blood form 25 per cent, of the same, the half of 

 which is hemoglobin, so a diminution in the solids means also a pro- 

 portionate diminution in the hemoglobin. Upon this hypothesis Ham- 

 merschlag has worked out a table 3 which represents the amount of hemo- 



3 hammkrschlag's table. 

 Specific Gravity. Hemoglobin. 



1034-1037 20-30 per cent. 



1037-1041 30-40 per cent. 



1041-1045 40-50 per cent. 



1045-1049 50-60 per cent. 



1049-1052 60-70 per cent. 



1052-1056 70-80 per cent. 



1056-1059 80-90 per cent. 



1059-1062 90-100 per cent. 



globin according to the specific gravity. I consider this a very convenient 

 method to use at the home of the patient, or when the hemometer cannot be 

 used. It is also fairly reliable for clinical purposes. 



The hematocrit precipitates the corpuscles in a graduated tube by 

 means of centrif ugalization, the bulk of the precipitate indicating the 

 number of corpuscles. With normal blood the corpuscles pack in the 

 tube, relatively to the number, but even then we do not know the elas- 

 ticity or compressibility of the normal corpuscle; so with different veloci- 



