246 HAEMOGLOBIN. 
added to a litre of saturated aqueous solution of crystals of oxyhemoglobin, 
and after the conversion into methemoglobin has occurred, about 25 per cent. 
of alcohol added. The mixture is then exposed to a temperature below 0° C. 
I succeeded in recrystallising methsemoglobin prepared by the action of 
potassium nitrite and of ethyl and amyl nitrites on oxyhemoglobin. 
Chemical and physical characters. — Crystals of methsemoglobin are 
more sparingly soluble than those of oxyhemoglobin, and the colorific 
intensity of their solutions is less. 
It is to be noted that, whilst solutions of reduced and oxyhemoglobin 
are not precipitated by either neutral or basic lead acetates, these 
reagents added cautiously, with careful avoidance of an excess, precipi- 
tate methoemoglobin, ho matin, and hcemaioporphyrin, and may be em- 
ployed for the separation and detection of traces of oxyhemoglobin 
when mixed with and concealed by any of the above-mentioned bodies. 
Solutions of methsemoglobin, when of a neutral or a slightly acid 
reaction, possess a chocolate-brown colour. When the solution is 
rendered alkaline, its colour changes to red without a tinge of the 
chocolate-brown. 
The acid solution is found to present a spectrum in which the oxy- 
hemoglobin bands a and j3 are very weak or even not visible, whilst an 
absorption-band is seen in the red between C and 1), and nearer the 
former. This band occupies nearly, though by no means exactly, the 
position of a similar band in the spectrum of acid heniatin (see Plate 
II., Spectrum 5). 
On now rendering the solution alkaline by means of ammonia, the 
band in the red disappears, and is replaced by a faint absorption-band 
immediately on the red side of D. By changing the reaction of the 
solution, the alterations in its colour and spectrum may be repeated 
indefinitely (Gaingee). 
If a solution of niethernoglobin be placed in a deep test tube, in 
front of a spectroscope, and arrangements be made for allowing a stream 
of solution of ammonium sulphide to flow to the bottom of the 
liquid, it can be readily shown that at the very moment of the 
contact of the reducing and the methaemoglobin solution, the spec- 
trum of oxyhemoglobin appears ; to be subsequently and much more 
slowly replaced by that of reduced hemoglobin, which in its turn, 
when shaken with air, yields oxyhemoglobin. 
A study of the photographic spectrum of methsemoglobin has led 
me to results of great interest. The conversion of oxyhemoglobin into 
methsemoglobin is attended by a shifting of the band of Soret from the 
extreme violet to the ultra-violet properly so called (Fig. 36). The most 
persistent part of the band in very dilute solutions, coincides, indeed, 
with the H and K bands, but the band extends more and more into the 
ultra-violet, as the concentration of the solution increases. The position 
and characters of this band in the case of methemoglobin absolutely 
corresponds with those of the acid compounds of hematin, and not 
with those presented by hemoglobin and its compounds, or by 
hemochromogen (see Fig. 38). 
This spectroscopic character certainly seems to lend weight to the 
evidence of other kinds, which indicates that methemoglobin is a first 
product of the decomposition of the oxyhemoglobin molecule, and that 
this is a decomposition which leads to the separation of a compound of 
hematin, and not of hemochromogen. Hoppe-Seyler, indeed, expressed 
