M. Preyer on the Colouring -matter of the Blood. 447 



length the inconveniences and sources of error to which both 

 these methods are liable. 



The method which he proposes as being more convenient and 

 more accurate depends on the following principle : — Concen- 

 trated solutions of haemoglobin e of a certain thickness, even 

 with a strong illumination, are opake for all rays with the excep- 

 tion of the red; while less concentrated solutions of the same 

 thickness allow a portion of green to pass along with red and 

 orange. Hence, if a measured quantity of blood placed in front 

 of the spectrum-apparatus be diluted with water until green 

 appears in the spectrum, and if once for all the strength of 

 hsemoglobine solution which under the same circumstances 

 transmits green be known, we can easily calculate the percent- 

 age of hsemoglobine in any blood. If k be the percentage in the 

 hsemoglobine solution, w the water added, b the measured vo- 

 lume of blood, x, the percentage of hsemoglobine in the blood, is 



_k(w + b) 

 ~ ~" T~ ; 

 or, if b— 0-500 cubic centim., 



x=k(l+2w). 



This method requires for its execution a spectrum-apparatus, 

 a finely-graduated accurately-calibrated burette, and also a pi- 

 pette by which T ^y cubic centim. can be read off, and a constant 

 source of light. Besides this, k must once for all be determined 

 experimentally. This is effected by preparing a solution of hse- 

 moglobine which, for a given spectrum-apparatus, a given un- 

 changed thickness of liquid layer (1 centimetre), and a given fixed 

 distance from the slit in the spectrum, just allows green to pass — 

 and in such a manner that the slighest increase of strength ex- 

 tinguishes it, while the slightest diminution increases the inten- 

 sity and then the breadth of the green layer. 



When k has once been accurately determined by drying the 

 solution over sulphuric acid in vacuo at 100° and weigh- 

 ing the residue, it is merely necessary that in each individual 

 experiment nothing be changed. The thickness of the layer of 

 liquid, the width of the slit, the distance of the unaltered source 

 of light and of the hsematinometer from the slit, and the position 

 of the prism must be constant in each experiment. To be inde- 

 pendent of changes of weather, a highly uniform petroleum- 

 lamp is used. 



The author describes the details of the mode of experimenting, 

 and the few precautions which require to be taken. He adduces 

 also a long series of analyses made by this method, and compares 

 the results with those which he obtained specially by determining 

 the amount of iron, and which show the accuracy of the method. 



