THE 
Use of the Spectroscope 
IN ITS APPLICATION TO SCIENTIFIC AND 
PRACTICAL MEDICINE 
EMIL ROSENBERG, M.D. 
m 
BY 
(The Essay, to which was awarded the Stevens’ Triennial Prize 
FOR 1876, BY THE COLLEGE OF PHYSICIANS AND SURGEONS, 
March ist, 1876) 
WITH ILLUSTRATIONS 
G. P. PUTNAM’S SONS 
NEW YORK 
182 Fifth Avenue 
1876. CONTENTS. 
CHAPTER I. 
The Apparatus and Method of Use I 
PAGE 
CHAPTER 11. 
Emission-Spectra and their Practical Application to Medicine , . 12 
CHAPTER 111. 
Absorption-Spectra. Blood 20 
CHAPTER IV. 
Practical Application of the Blood-Spectrum . . 29 
The Absorption-Bands of Bile and Urine 37 
CHAPTER V. 
CHAPTER VI. 
Spectrum Analysis of the Blood after introduction (inhalation) of 
foreign gases 42 
Appendix 62 DESCRIPTION OF THE TABLE 
I. Solar Spectrum. 
2. Chloride of Cobalt solution (in absolute alcohol). 
3. Blue glass. 
4. Chrom-alum (sol. in water). 
5. Ammonio-sulphate of copper (in water). 
6. Prussian blue (in oxalic acid). 
7. Indigo (in sulph. acid). 
8. Chloride of copper (in water). 
9. Permanganate of potassium. 
10. Aniline-Red. 
11. Bichrom. of potassium (in water). 
12. Green glass. 
13. Bergamot-oil. 
14. Chlorophyll (in ether). 
15. Tinctura Aconiti ex foliis. 
16. Carmine (ammoniacal solution). 
17. Oxyhaemoglobin. 
18. Haemoglobin (Reduction-band). 
19. Haematin (Oxyhaematin). 
20. Haemin (Acid-band). 
21. Reduced Haematin. 
22. Methaemoglobin. 
23. Blood treated by SNa in excess. 
24. Fel tauri inspissatum (in muriatic acid). 
25. Urobilin, acid solution. 
26. Urobilin, alkaline solution. TABLE OF SPECTRA (FLINT PRISM OF 6o°). THE USE OF THE SPECTROSCOPE. 
CHAPTER I. 
THE APPARATUS AND METHOD OF USE. 
It is not proposed, in this brief essay, to give a complete 
exposition of all the optical relations of our subject, as this 
want has been thoroughly met in the masterly productions 
of Schellen, Roscoe, and Lockyer, who have fully and satis- 
factorily established themselves as the authorized and ac- 
knowledged exponents of the general principles of Spec- 
trum Analysis. But although in these works the appara- 
tus, with all its varieties, is fully described, we think it 
necessary to reproduce here a detailed description of the 
instrument and the mode of using it. 
The principal part of the Spectroscope (that by which 
the decomposition of light is effected) is either a grating 
or a prism. We shall confine ourselves to the discussion 
of the prism, which has led to such extensive discoveries; 
though it must be conceded that it is only by means of 
the diffraction-grating that a truly normal spectrum can be 
obtained. 
The prisms used are either solid or hollow, the latter be- 
ing filled with a fluid of high refractive power (bisulphide 
of carbon, CSg). One or more prisms may be employed at 2 
THE USE OF THE SPECTROSCOPE. 
the same time. The single prism may be either simple or 
compound. The simple glass (dense flint, of a refracting 
angle of 6o°), or the ordinary CSs prism, effect a deflection 
(deviation) and dispersion of the incident ray, allowing only 
oblique vision. The compound glass prism (a combination 
of several flint and crown glass prisms) enables the refracted 
ray to emerge in the straight continuation of the incident 
beam. These compound prisms, therefore, permit direct 
vision, the spectrum being seen by looking straight at the 
source of light. During the last two years we have been in 
possession of a compound CSs prism. The inventor, Prof. 
A. K. Eaton, of Brooklyn, N. Y., conceived the idea of 
combining the CSg prism with a piece of crown glass, which 
combination, of course, also permits direct vision. 
Although a prism is, in a certain sense, a Spectroscope 
per se, there are, besides, in the construction of a complete 
apparatus, several other essential parts : 
Fig. i. 
i. A vertical slit, the breadth of which can be regulated by means of a screw, at the 
end (a) of the collimating tube {A) ; at the other end, nearest the prism, an 
achromatic lens (the Collimator) is placed at the distance of its focus ■ for rays 
from the slit, which are made parallel by the Collimator and thrown upon the 
prism. THE USE OF THE SPECTROSCOPE. 
3 
2. An astronomical telescope (B) of low magnifying- power, through which the rays 
of light, refracted and decomposed into a spectrum, reach the eye of the observer. 
3. For fixing the position of spectral lines or bands, various auxiliary contrivances 
are employed ; thus, for instance, Kirchhoff and Bunsen have introduced, at the 
head of a third tube (c), a reduced photographed millimetre scale (/;), which can 
be seen through the telescope by reflection from the front surface of the prism. 
The three tubes are fastened to a central plate, carrying the prism, and screwed 
to the upper end of a cast-iron foot. The Collimator-tube (or the arm support- 
ing it) must be immovable. The telescope and scale-tubes are so fitted that they 
are movable in a horizontal plane about the axis of the base. In front of the slit 
is a small equilateral glass prism, covering the lower half of the slit. This prism 
of comparison transmits by total reflection the light of a second lateral source 
through the slit, so that the observer sees the spectra of the two sources of light, 
one under the other. 
The arrangement of such an (Bunsen and Kirchhoff) 
apparatus is effected in the following manner : 
The prism is taken off the stand, and the observer looks 
first through the collimator-tube in the direction b to a, 
with moderately opened slit, the slit-tube being drawn out, 
until the edges of the slit appear sharply defined. Then, 
the telescope, having been adjusted for a distant object, is 
brought into position in a straight line with the collimator- 
tube, and the slit observed through the telescope ; the lat- 
ter has to be moved, so that the slit will appear as a sharp 
bright line in the middle of the field of view. The prism is 
now brought again to its place. The scale-tube is turned 
through an angle round the axis of the base, and the illu- 
minated scale is pushed in with its tube, until the division 
of the scale, observed through the telescope, is distinctly 
perceived. Stray light is excluded by covering the prism 
and the ends of the tubes adjoining it with a black cloth. 
The illumination of the scale is effected by means of a 
light placed before it. A white paper screen interposed 
between the flame and the scale will obviate the unpleasant 
glare; and often the reflex of the diffused light of the 
room from a white paper will be found sufficient for the 
illumination of the scale without an additional light. 4 
THE USE OF THE SPECTROSCOPE. 
Other apparatus* differ from the instrument just de- 
scribed in being provided with only two tubes; the scale 
being placed in the telescope, just behind the ocular. 
The central plate may easily be arranged for prisms of 
different kinds. When a simple glass or CSg prism is em- 
ployed, the telescope will have to be passed through a 
certain angle round the ,axis of the base. The compound 
glass, or Eaton’s compound CS3 prism will, on the contrary, 
require both tubes (the telescope and the collimating-tube) 
to be in a nearly straight line. In this case we should 
have a direct-vision Spectroscope. 
The prism has to be placed so that the edge of the re- 
fracting angle is parallel to the slit. Besides, the position 
of the prism depends upon the degree of the dispersion 
required ; and, in order to obtain a pure spectrum, it must 
be that of “ the angle of minimum deviation.” 
For the fuller explanation of the latter point, we refer 
to the subjoined diagram, 
Fig. 2. 
i in which a b c represents a prism, a b and a c its refracting 
faces, d e the incident ray, which is refracted within the 
prism as ef, and then again emerging as fg. The recip- 
* The instruments made here, in New York, have a mirror attached to the 
Collimator, with two movements for reflecting the rays of the suit into the slit. THE USE OF THE SPECTROSCOPE. 
5 
rpcal direction of the incident and the emerging ray de- 
pends (other conditions being equal) upon the angle of 
incidence <p (formed by d e with he, the perpendicular 
upon a b). 
This angle varies with the position of the prism. By 
turning the prism around the axis of its length, a position 
will be obtained, in which the angle of incidence will be 
equal to the angle of refraction {d eh g f I), and in which 
the angle {in kg) formed by the prolongation of the emerg- 
ing and of the original path of the incident ray within the 
prism, acquires its minimum value. The angle mk g, mea- 
suring the amount of deviation of the prismatic image, 
shows the deflection to be reduced to a minimum, and is 
therefore called the angle of minimum deviation. 
Practically, the position of minimum deviation is found 
by slowly turning the prism round its axis, until the Fraun- 
hofer’s lines or the bright metal lines will exhibit the least 
perceptible change as to their location. Besides it is the 
precision * with which certain definite lines are perceived— 
for instance, the Fraunhofer’s lines a, B, C, that indicates 
whether the minimal deviation is in accordance with their 
respective region—in this instance with the Red. From all 
this it would follow, that the simple prism has to be brought 
into various positions, according to each separate line which 
we wish to observe. But in practice this will be found to 
be unnecessary. 
In addition to the ordinary numbered scale, there are 
other methods in use for the localization of spectrum lines 
or bands. 
A well defined mark of some kind, such as a wire or 
* In addition to all the above-described procedures, it is usually necessary to turn 
the whole instrument slightly around the axis of its length, in order to obtain Fraun 
hofer’s or metal lines with greatest precision. 6 
THE USE OF THE SPECTROSCOPE. 
cross-wires, or a line of light, are made to traverse the spec- 
trum ; and the amount of motion is measured externally by 
means of a micrometrical arrangement ; or the position of 
that part of the spectrum covered by the cross-wires may be 
read off a horizontal graduated circle provided for that 
purpose. 
Another method for fixing the location of appearances 
in the spectrum has more recently been devised by Profes- 
sor J. C. Dalton.* Its chief feature is a photographed scale, 
presenting not only divisions, but the position of the Fraun- 
hofer’s lines permanently marked—of course for a certain 
permanently fixed position of the several parts of the 
instrument. 
Among the varieties of Spectrum Apparatus, there is 
one of great importance for the physiologist and physician ; 
viz. the micro-spectroscope, by which the absorption-phenom- 
ena of the most minute solid and liquid bodies, for instance, 
that of a corpuscle of blood, may be accurately observed. 
The spectroscopical part of such an instrument (Sorby- 
Browning) can be applied to any microscope by fixing it in 
the place of the ordinary eye-piece. A complete description 
of the micro-spectroscope may be found in the works of 
Schellen or Roscoe. 
Before commencing practical operations with the Spec- 
troscope, it is necessary to determine—in relation to the 
scale employed—the position of the principal Fraunhofer’s 
lines f (A to H), as they appear in the sun spectrum, if a 
narrow*slit is used. (With a widely opened slit, the sun will 
* Dalton, J. C., A new method of determining the position of absorption-bands 
in the spectrum of colored organic fluids. (A paper read before the N. Y. Acad 
emy of Medicine, in October, 1874.) 
t See note at the end of this chapter. THE USE OF THE SPECTROSCOPE. 
7 
produce a continuous spectrum, as Newton saw it). A de- 
scription of the scale thus formed must precede every report 
of spectral observations, for the reason that each spectrum, 
obtained by prismatic refraction, is always only an individ- 
ual one. Such a “scale ” may be chosen once for all and 
retained. 
The simplest method .will be to determine all the prin- 
cipal Fraunhofer’s lines by direct observation in sunlight; 
the position of the prism and its distance from the collimat- 
ing lens being fixed. Both may be easily marked on the 
instrument, or made permanent by special contrivances. 
Even when there is but little sunlight, the lines D, E, b, and 
F are determinable by direct observation ; the others are 
readily calculated by Kirchhoff’s or any other exact scale. 
The Fraunhofer’s lines may, on the other hand, be deter- 
mined independently of the sun. Thus Heynsius and 
Campbell * have adopted the following rather complicated 
method. In the first place D is fixed by means of the 
Sodium line Na, coincident with D ; and A through the 
dark red Potassium line Ka, coincident with A. Then the 
position of C and F is found by means of a’Geissler tube 
filled with hydrogen gas.f After that, the lines a, B, E, and 
b have to be calculated according to “ Schellen’s scale.” 
Occasionally, the scale is made only with reference to 
some metal lines. Thus the scale chosen may be sufficiently 
defined by mentioning the position of the lines K a , Li a, 
Na, Sr S. 
The employment of metal lines is absolutely indispens- 
able in controlling and verifying the position chosen in day- 
f Electric sparks passing through such a tube, containing hydrogen at the pressure 
of one atmosphere, produce a bright carmine red of the incandescent hydrogen, which 
color is resolved by the spectrum into a red line (coincident with C), a bluish-green 
line (coincident with F), and a blue line (coincident with G). 
* Archiv f. Physiologic, iv. 1871, p. 520. 8 
THE USE OF THE SPECTROSCOPE. 
light, when artificial light is used. The bright yellow 
Sodium line is especially employed for this purpose. 
The scale presented below is obtained 
by a simple flint prism of 6o°, used by the 
author; the dispersion of which in a cer- 
tain position being so small as to allow the 
entire spectrum to be brought into the com- 
pass of the scale : 
Fig. 3. 
The red end of the spectrum is brought to the observer’s left 
hand, by placing the refracting edge of the prism to the 
left. The spectrum observed through the prism without 
the telescope, presents in this position the red end to the 
right—through inversion by the telescope it is seen at the 
left. 
When a greater dispersion is required, 
or the spectrum is displaced towards the 
violet end, in order to bring the red end 
nearer to the centre of the field of view, 
only a portion of the spectrum will be visi- 
ble at one time ; the remaining sections can 
only be brought into view by moving the 
telescope horizontally. When artificial light 
is used, one of the metal lines occurring in 
the blue section of the spectrum, for in- 
stance, the blue Strontium line, Sr S, or the 
blue Caesium line, Cs a, can be employed 
for fixing F and G, provided the distance of 
such a metal line from both Fraunhofer’s 
lines has been determined in daylight, and 
the relative position of the several parts of the instrument 
retained. If the dispersion is not too great, the Thallium 
line, or even the Na line may be employed for the same 
purpose. THE USE OF THE SPECTROSCOPE. 
9 
(In the three-tube instrument, the Collimator-tube, the 
prism and the scale-tube are the three parts, the reciprocal 
position of which has to remain unaltered, if the once- 
chosen scale is to be retained.) 
In reference to the calculation previously alluded to, we 
insert in this place Kirchhoff and Bunsen’s Reduced and 
Larger scales: 
KIRCHHOFF’S LAR.GE SCALE. 
kirchhoff’s reduced scale. 
A = 405 
a = 505 
B = 593 
C = 694 
D, = 1002.8 
D 2 j = 1006.8 
E = 1522.7 
b, ) = 1633.4 
b2 V = 1648.3 
b3 ) = 1655.0 
F = 2080 
G = 2855 
h 3371 
H - 3568 
A = 17.5 
a = 23 
B = 28 
C = 34 
D = 50 
E = 71 
b = 75.9 
F = 89.8 
G = 127.5 
h 148 
H, = 161.9 
H2 =*l66 
Preyer’s scale (“ Die Blutkrystalle,” p. 227,) may also be 
profitably employed for this purpose : 
preyer’s scale. 
A = 31.2 D = 59.5 
a = 36.5 E = 77.9 
B = 41.0 b = 81.3 (b, = 81.0, and b4 = 81.6) 
C = 45-9 F = 94-6 
G = 128.5 (128.0—129,0) 
Hj = 158.7 (158.2—159.2) 
The measures refer to the centre of the lines or groups. 10 
THE USE OF THE SPECTROSCOPE. 
Though the calculation is very simple, it may not be 
out of place to illustrate it by an example: 
Suppose D, E, b and F having been observed by sun- 
light, A would be found by this rule: 
* The distance DE (our scale) : x (the distance AD) = 
the distance DE (Kirchhoff) : the distance AD (Kirchhoff.) 
In order to attain the necessary exactness in using the 
bright lines as substitutes for Fraunhofer’s lines, the following 
point has to be considered. The bright spectrum lines will 
always show the width of the slit. In widening the slit by 
means of the screw only one edge is displaced, the other 
being immovable. If the latter be to the right hand of the 
prism, the left edge of the bright line will—through the in- 
version by the telescope—remain at its place, while the ex- 
pansion of the bright band will take place towards the right. 
We would therefore have, in this case, to bring the left 
edge (for instance, of the yellow Sodium line) under the 
number of the scale chosen for the line D. 
Besides the vertical Fraunhofer’s lines, the observer will 
frequently see horizontal lines. These are produced by 
inequalities of the slit or by dust. By cleansing the slit 
with a small brush or piece of paper they will necessarily in 
the latter instance disappear. They are, however, useful in 
adjusting the distance of the slit from the prism, which is 
only correct, if the vertical Fraunhofer’s lines, the horizontal 
dust lines, and the divisions of the projected scale (or the 
cross-wires) are perceived with equal distinctness. It is 
evident that these dust lines will be even more useful in 
this respect in the absence of Fraunhofer’s lines, viz., when 
artificial light is employed. THE USE OF THE SPECTROSCOPE. 
THE FRAUNHOFER’S LINES. 
Each glowing gas * absorbs exclusively rays of the same 
refrangibility as those which it would itself emit (radiate) in 
the state of incandescence. 
The Emission-(radiation-) spectrum of each gas glow- 
ing in the solar atmosphere, for instance of hydrogen, 
which consists of the three bright lines already mentioned, 
must be reversed (appear black) if rays are transmitted 
through it from a source of light sufficiently bright, and 
giving for itself a continuous spectrum (the white-hot 
central body of the sun). 
* Metal molecules glowing in flames or in the electric spark, and producing 
thereby the bright metal lines, are also in gas form. CHAPTER 11. 
EMISSION-SPECTRA, AND THEIR PRACTICAL APPLICATION 
TO MEDICINE. 
Solid and fluid substances emit in the state of incan- 
descence, light of every degree of refrangibility; that is, 
they give a continuous spectrum.* An exception to this 
law is presented by solid Erbia, which exhibits bright 
lines. 
Each body emits in the condition of glowing gas, light 
of various degrees of refrangibility, giving a discontinuous 
spectrum which contain bright bands or lines upon a dark 
background. 
The incandescent vapors of the alkalies and alkaline 
earths; Sodium, Potassium, Lithium, Calcium, Strontium, 
Barium, Rubidium, Caesium, Thallium, and Indium, give 
discontinuous spectra, for the most part visible by means 
of the spirit lamp, or through the non-luminous flame of the 
Bunsen burner (or the hydrogen flame). The heavy metals 
(metals proper) require the heat of the electric spark for 
their volatilization. 
We have seen that a knowledge of some of the bright 
metal lines is necessary, if only for the determination of the 
Fraunhofer’s lines when artificial light is used. 
* Law discovered by J. W. Draper ; “In view of the foregoing facts, I conclude 
that, as the temperature of an incandescent body rises, it emits rays of light of an 
increasing refrangibility.” (On the production of Light by Heat. Phil. Mag. xxx. 
iB47> P- 345•) THE USE* OF THE SPECTROSCOPE. 
13 
In medicine, the Emission-Spectrum Analysis is applied 
to the establishment, in animal (or vegetable) tissues, of the 
normal or accidental occurrence of chemical elements, which 
might not be demonstrable by ordinary chemical analysis 
with the same facility, or the identification of which might 
be impossible, from the exceedingly small quantity in which 
they may be present. 
For the purpose of examination, the animal tissues are 
brought into the flame, either fresh, so as to be carbonized 
* 
and partly incinerated, or they may first be reduced to ashes 
(in the platinum crucible) and thus examined without fur- 
ther preparation. Another method consists in treating the 
coal or ashes by pure muriatic acid, so as to transform the 
alkalies into their chlorides, which are volatilized with 
greater facility. The details of the procedure are as 
follows; 
The end of a thin wire of platinum is looped, apd heated 
until it ceases to give a luminous flame; a bead is then 
melted into the loop by taking up a small quantity of the 
ashes with the moistened wire, and drying it at the surface 
of the flame. The Bunsen burner is then placed before the 
slit at a distance of about 5 to 10 centim., the upper edge 
of the chimney being 2 or 3 centim. beneath the lower end 
of the slit. The air is shut off below, so as to permit a 
brilliant spectrum to appear through the telescope, which 
determines the right position for the burner. The air is 
then passed through the openings into the mixing-chamber, 
the scale is illuminated, and the platinum wire, fastened to 
a glass tube, is brought into the outer mantle of the non- 
luminous flame. 
In observing the spectra of alkaline chlorides in general 
(from solutions), the contrivance known as “ Mitscherlich’s 14 
THE USE OF THE SPECTROSCOPE. 
tube or wick” will be found much more convenient than 
the thin wire.* 
We propose now to give a description of the character- 
istic features of these spectra; referring, for more specific 
details, to the abstract and drawings, after Bunsen and 
Kirchhoff, given in Roscoe’s work.f 
Sodium (Na) occurs in the animal economy, more especially 
in bloodplasma, urine, pancreatic fluid, the bile of man 
and the majority of animals, and also in serous transu- 
dations. It imparts to the non-luminous flame a bright 
yellow color, and presents a spectrum consisting of only 
one yellow line, which is coincident with Fraunhofer’s 
dark line D. (These notes refer only to the spectra of the 
chlorides, as obtained by means of the Bunsen burner.) 
Potassium (K) is more especially present in the ashes of 
muscle, of red blood corpuscles, nerves, yolk of egg, 
and milk. It gives a pale violet color to flame, and 
the Spectrum presents two characteristic lines : 
Ka|, dark red, coincident with the dark line A of the 
solar spectrum. § 
K/3, violet, very near H. 
The Potassium lines are often concealed by the inten- 
sity of the yellow line of Sodium, which mostly appears 
* The author finds ordinary steel-pens a good substitute for the wire. The point 
being fixed in the holder, the pen forms a convenient metal-spoon for subjecting traces 
of the chlorides to the heat of the Bunsen burner. 
t Kirchhoff and Bunsen, “ Chemical Analysis by Spectrum Observations.” First 
Memoir : Pogg. Ann. CX. i6x. Phil. Mag., 4th series (i860) xx. Bg. Second.Memoir : 
Phil. Mag. xxii. xB6i. Abstract in Roscoe, 1869, p. 68. 
t The Greek letters designate the intensity, a signifying the brightest line, 
• § M. Louis Grandeau (Instruction pratique sur I’Analyse Spectrale, Paris, 1863—a 
book which is to be highly recommended) declares on the contrary the line Ka not to 
be coincident with any Fraunhofer’s line. Father is correct. Ka coincides with A, if 
the observation is made with a simple apparatus and with a telescope of low magni- 
fying power. It appears less refrangible than A, if observed by the electric spark, 
several prisms and high magnifying power (Kirchhoff). THE USE OF THE SPECTROSCOPE. 
15 
simultaneously. By interposing a cobalt solution (Ta- 
ble, Fig. 2), or a blue cobalt glass, the intense Sodium 
light may in such a case be weakened ; by interposing 
an Indigo solution (Fig. 7,) it will be entirely obscured. 
The Potassium spectrum has the peculiarity of being 
continuous in the middle part, from yellow to blue.* 
Lithium (Li) has been found by means of its spectrum in 
traces in muscle, blood, and milk of the human subject 
and animals. The flame is of a crimson color, and 
shows a spectrum consisting of a characteristic bright 
red line, Li a, between B and C, and a second very 
faint yellow line, Li /3, between C and D. 
Calcium (Ca) is largely found in bone and teeth ; in smaller 
quantity in any animal fluid, and many pathological 
products. The flame is of a yellowish-red color, and 
shows a spectrum consisting of nine lines, of which 
two are characteristic: 
Ca a a broad orange line between C and D ; 
Strontium (Sr) is introduced into the animal system by 
food or water containing it. It colors the flame red. 
The spectrum consists of eight lines, four of which are 
characteristic ; 
Ca f3, a broad bright green line between D and E. 
Sr a, Sr (3, Sr y, are red lines, between C and D, the 
first named close by D, or the Sodium line. 
Sr (5 is blue, between F and G. 
Barium (Ba) exceptionally occurs in the body, from the 
same sources as Strontium; it produces in flame a 
* The Potassium lines are very easily and conveniently perceived, by holding a 
grape-stem (tartrate of potassa) before the slit in the non-luminous flame of the Bun. 
sen burner. 
Matches, prepared with chlorate of potash, will likewise be found very useful in 
this respect. THE USE OF THE SPECTROSCOPE. 
green tint, and gives a spectrum consisting of fifteen 
lines, three of which are characteristic ; 
Ba a, a broad green band between E and b. 
Ba 13, near b, towrards the blue ; and 
Thallium (Tl) colors the flame green. The spectrum con- 
sists of one bright green line between D and E, nearer E. 
Bay, between D and E. 
Indium (In) shows a violet-colored flame. The spectrum 
consists of two lines: * 
In a, between F and G, blue. 
In (3, between G and H, violet. 
Rubidium (Rb) exhibits a reddish-blue flame. The spec- 
trum comprises ten lines, four of which are characteristic : 
Rb a, and Rb (3 are violet lines, between G and H. 
Rb y, and Rb S are red lines, which fall beyond 
Fraunhofer’s line A. 
Caesium * (Cs) presents a similar-colored flame, the spec- 
trum embracing eleven lines, three of which are char- 
acteristic : 
Cs a and Cs (3, blue lines between F and G 
Cs y, red, between C and D. 
There is no difficulty in producing a compound spectrum 
of K, Na, Li and Ca from the tissues of man or mammals. 
Every tissue which is rich in blood-vessels, will yield the 
lines of one of these metals, even if the latter be not con- 
tained in the tissue itself, but only in the blood. 
It is therefore obvious, that bloodless tissues, destined 
for Spectrum Analysis, must be separated at a perceptible 
distance from their matrix. 
Lithium has been found in parings of nails, and in the 
* Rubidium and Caesium possess the same peculiarity as Potassium, giving a 
continuous spectrum in the centre. THE USE OF THE SPECTROSCOPE. 
17 
hair of living man. Hence it occurs independently from 
any admixture of blood.* 
An application has been made by Professor Valentin,f 
of one of these metal lines to physiological research. He em- 
ployed the Strontium spectrum as a substitute for the chem- 
ical (Hering-Vierordt’s) ferrocyanide of potassium test, for 
the purpose of measuring the velocity of absorption or of 
the circulation. Sodium, Potassium, etc., being normal 
constituents of the system, can not be so employed. 
Strontium, however, being in exceptional cases intro- 
duced into the body, a previous test of its presence with a 
drop of the blood of the animal to be experimented upon is 
required. But even should its accidental presence in min- 
ute traces be proved, it is only necessary to use a weaker 
flame, which, under the circumstances, would not produce 
the spectrum. 
(For an application of the bright metal lines in Ophthalm- 
ology, see Appendix.) 
In reference to the demonstration of heavy metals, we 
find in Kirchhoffs tables the lines of Zinc (Zn), Nickel 
(Ni), Magnesium (Mg), Gold (Au), Tin (Sn), Mercury (Hg), 
Copper (Cu), Iron (Fe), Aluminium (AI), Arsenic (As), 
Stibium (Sb), Lead (Pb), Cobalt (Co), Silver (Ag), Chro- 
mium (Cr), and Cadmium (Cd). Among these are some of 
the most ordinary poisoning agents. But, although Valen- 
tin, as early as 1863, declared that “ the application of this 
means to the diagnosis of metal poisoning will undoubtedly 
* For spectral-analytical demonstration of T1 and Li in urine, see Neubauer und 
Vogel, Analyse des Hams, 6th edition, 1872, pp. 134 and 135. 
f Valentin, G. Der Gebrauch des Spaktroscopes zu physiologischen und arztlichen 
Zwecken. Leipzig, 1863. 
X Compare also Wm. Huggins’ maps of the metallic lines. These, as well as 
Kirchhoffs and Angstrom’s, are found in Roscoe. 18 
THE USE OF THE SPECTROSCOPE. 
form one of the brightest features in the medical use of the 
spectroscope,”—we must confess that there is, so far, not a 
single observation on record concerning the examination by 
the spectrum of animal tissues impregnated by such poisons, 
—not even an experimental one. The temperature of the 
Bunsen burner, notwithstanding its 2350° C., is not sufficient 
for the volatilization of the heavy metals (with the excep- 
tion of copper and manganese). Their conversion into in- 
candescent vapors must be effected by the electric spark, as 
produced by 20 to 40 Bunsen elements, charged with nitric 
acid, or with a weaker battery, by the aid of the Ruhm- 
korff Inductor, sometimes strengthened by the interpolation 
of a Leyden jar. The animal tissues should for this pur- 
pose be placed in a small, flat coal crucible, to which the 
spark from the other charcoal-point is made to pass over. 
As to the Emission-Spectra of the: gases and non- 
metallic elements (Air, Hydrogen, Nitrogen, Oxygen, Sul- 
phur, Selenium, Phosphorus, Chlorine, Bromine, lodine), we 
must confine ourselves to the mention of their production 
by electric discharge, through Geissler tubes filled with the 
rarefied gases. That these spectra are vastly important in 
spectroscopic study in general, has been elsewhere shown 
in reference to Hydrogen. 
Regarding their application in medicine, Valentin’s pro- 
position may be of interest to the student of comparative 
physiology. He says*: “These gas-spectra, especially 
those of hydrogen and carbonic acid, might be employed 
for qualitative Analysis of gases, existing only in minute 
quantity in several parts of the animal economy—as the 
intestinal gases of the frog, insects, and other creatures. 
The gases would have, however, before introducing them 
* Loc. cit., p. 131. THE USE OF THE SPECTROSCOPE. 
19 
into a Geissler tube, to' be dried thoroughly with sulphuric 
or phosphoric acid, and divested of Ammonia, if it be 
present; and, besides, the atmosphere would have to be 
rigidly excluded during the filling of the tube.” 
If Spectrum-Analysis has so far been successfully applied 
only to the discovery of the alkaline metals and earths, while 
the spectra of the metals proper, from their complicated 
nature and other difficulties, seem not destined to become 
of great practical value to the physiologist—it must be 
observed, that it is the former, that are with the greatest 
difficulty distinguished by ordinary chemical reactions, 
especially if—as is generally the case—only present in 
minute quantity in a mixture. CHAPTER 111. 
ABSORPTION SPECTRA.—BLOOD. 
NOTWITHSTANDING the important facts offered in medi- 
cal Spectrum Analysis by the Emission phenomena, yet the 
main subject for our consideration is that section relating to 
spectral appearances of transparent colqred bodies, more es- 
pecially fluids.* 
If sunlight, or the light of a bright flame be passed ver- 
tically through a colored solution, placed immediately be- 
fore the slit in a glass vessel with plane-parallel walls, a 
great number of pigments will, by the gradual dilution of 
the fluid, present a discontinuous absorptive spectrum, 
certain definite sections thereof being strongly absorbed, 
while the adjacent sections either remain bright, or the 
absorption is much less vigorous. With certain dilutions, 
there will appear one or more-narrow or broader absorption- 
bands, the position and extent of which are definable by 
means of the scale in relation to the Fraunhofer’s lines. 
The position of absorption-bands is either defined simply 
by the corresponding divisions of the graduated scale, adding, 
if necessary, the Fraunhofer’s lines between which they are 
located ; for instance a certain band may be said to have the 
position of 53-57, between D and E. 
Or the following method introduced by Valentin may 
* Gladstone, J. H., on the use of the prism in qualitative analysis. Chem. Soc. 
Journal, 1858. (Absorption Spectra of various colored metallic sale solutions.) THE USE OF THE SPECTROSCOPE. 
21 
be adopted. Suppose there be 19 divisions between D and 
E, a band may be thus described : 
E to E. 
A band commencing at A, and extending midway be- 
tween B and C, is described as a band from A to B—C. 
The location of bands by this latter method is verified with 
greater facility by other observers than the determination 
by the former plan, which would require calculations similar 
to those referring to the Fraunhofer’s lines (p. 10). 
These absorption spectra must not be estimated at the 
same value as the flame spectra of the metals. The most 
valuable element of these latter is, that in a mixture of 
metal compounds, the individual metals may be discovered 
and distinguished with so great a facility; and this can 
scarcely ever be attained in regard to pigment mix- 
tures. 
Further, the absorption phenomena, consisting merely 
in the weakening or entire extinction of several parts of the 
spectrum, will rarely offer the sharp outlines of the bright 
metal lines,—the transitions from the absorbed.to the un- 
altered part of the spectrum (the limit of absorption) being 
seldom definable with precision. These appearances are 
also not only dependent upon the pigment itself, but upon 
a number of other incidents, as the thickness of the layer of 
the fluid, the degree of concentration, temperature, light- 
intensity of the spectrum, admission or exclusion of air, etc. 
We now propose to discuss the Spectrum appearances 
of the colored fluid, which preeminently claims our attention. 
A valuable generalization is contained in the following 
sentence : 
“ Each body appears dark in the light, which it is incom- 22 
THE USE OE THE SPECTROSCOPE. 
petent to transmit.” Applying this to the blood-pigment,* 
we may say ; 
Hemoglobin (Hb)f, when combined with oxygen; 
appears dark at two distinct points of the yellow-green, 
between the Fraunhofer’s lines D and E of the sun- 
spectrum. 
Hoppe-Seyler, in 1862, taught this fact," in regard both 
to the Oxyhemoglobin [Scarlet Cruorin, Stokes] crystals 
(00-Hb), and to their solutions—and also to blood-solutions. 
a Virchow’s Archiv 23, 1862. Hoppe-Seyler (first communication). 
“ “ 29,1864. “ “ '(2 d and 3d communication). 
The most important paper on the subject of the physiologo-chemical 
(including spectrum) relations of Haemoglobin and its products of 
decomposition is : Hoppe-Seyler, Beitrage zur Kenntniss des Blutes 
des Menschen und der Wirbelthiere, in Med. chem. Untersuchungen. 
Berlin, 1866-1871. 
He describes the gradual development of the Og-Hb 
spectrum in about the following way: 
A concentrated Haemoglobin solution allows but a por- 
tion of the red to pass through. Diluted with water, it 
shows, when examined by sunlight, a rapid clearing up to 
D ; light speedily appears between E and F, in the green ; 
and, after further dilution, the spectrum expands over F, 
while, at the same time, a bright yellow-green band is seen 
midway between D and E, leaving two dark bands in this 
space. On continued dilution, the entire spectrum will 
gradually appear up to the violet, and the two absorption- 
bands alone remain distinctly visible at a dilution of I grm. 
* Kuehne found the red pigment of muscle to be identical with Haemoglobin (inde- 
pendent of the blood therein contained). E. Ray Lankester (Pflueger’s Archiv iv. 
1S71) confirmed this in reference to particular muscles of various animals, especially 
the muscular fibres of the pharynx in certain mollusks, the blood of which does 
not contain any Hb. 
f Syn ; Haematoglobulin ; Cruorin (Stokes). THE USE OF THE SPECTROSCOPE. 
23 
Haemoglobin to 10,000 C.C. fluid, and a thickness of the 
solution of 1 Cm.* 
The first band (a), nearer to D, is darker, narrower, and 
more sharply defined than the second (/3), which nearly 
approaches E; and upon increased dilution, the first band 
will at last disappear, but somewhat later than the second 
band. 
Ck-Hb is deprived of its oxygen by setting aside a 
sufficiently diluted blood-solution in a hermetically-sealed 
vial, or by raising such a solution, in a closed, vessel, for a 
short period, to the temperature of the body,—or by add- 
ing reducing agents, for instance, one or several drops of 
liquid hydrosulphate of ammonia. The arterial color will 
then gradually fade, and the examination per spectrum pre- 
sent the bright interval between the two bands obscured, 
the bands themselves becoming fainter, and being replaced 
at last by the so-called reduction-hand, a broad, imperfectly- 
defined absorption-band,—the darkest field of which is cen- 
tral between D and E. At the same time the blue part of 
the spectrum appears much less absorbed than in an 03- 
Hb solution of equal concentration. If such a solution of 
reduced Hb (purple Cruorin, Stokes) be shaken up only 
for a short time with oxygen or air, the two bands will 
instantly reappear, to vanish and again give place to the 
reduction-band, if and so long as the reducing substance 
be present in the liquid. The reduction-band was discov- 
ered in 1864, by Stokes.f 
Besides hydrosulphate of ammonia, other substances, 
which have a marked faculty of reduction in alkaline 
* Of fresh defibrinated blood the proper dilution is i in 100 (vol.) water, to be 
viewed in a layer of i centimetre. 
f George Gabriel Stokes, the celebrated occupant of Newton’s chair, in Cam- 
bridge. (Philosophical Magazine, 1864.) 24 
THE USE OF THE SPECTROSCOPE. 
liquids, are employed. Among them are an ammoniacal 
solution of the tartrate of protoxide of iron,—an ammon. 
sol. of tartrate of protoxide of tin,—and sulphide of sodium 
(SNa), the latter warmly recommended by Preyer.* Of 
these reagents, the Tin solution offers certain special ad- 
vantages. It is inodorous, colorless, and effects (even when 
added in larger quantity) simply reduction, while sulphide 
of ammonium and SNa decompose the blood, when em- 
ployed in excess (Nawrocki). A blood-solution treated by 
protoxide of tin may be kept for weeks in open vessels, 
without undergoing the least change. 
The tin-liquid is thus prepared : An aqueous solution 
of the common protochloride of tin (tin-salt) is treated by 
tartaric acid and neutralized by ammonia. The acid has to 
be added in such a quantity, that, after over-saturation by 
ammonia, no precipitate results, leaving a perfectly clear 
solution. 
We are thus enabled to distinguish through the spectrum 
the presence and absence of oxygen in blood, because the 
blood is in each case of a different color. Alterations of the 
blood to be discernible by absorption-spectrum analysis, 
must include the formation of new pigments, which need 
scarcely be discernible to the unaided eye. 
It will suffice here to refer to the close analogy between 
the artificial reduction and the change which the arterial 
blood undergoes in the living organism in its transformation 
into venous blood. 
Beside the reduction, we have the decomposition of the 
bloodred, which by its products becomes the object of Spec- 
trum Analysis. Of these products (albumines, pigments, or 
acids), the pigments alone affect the present question. 
* Preyer, W., Die Blutkrystalle, Jena, 1871. THE USE OF THE SPECTROSCOPE. 
25 
Several of these are artificial products of decomposition, 
which also occur in the living body. The most important 
of this class is Hcematin. 
Hb being a colored albuminous compound, nothing is 
required to effect its decomposition but the means ordinarily 
used for coagulating or precipitating albumen ; viz., heat, 
alcohol, mineral acids, organic acids (even the weakest), and 
strong alkalies. 
Hcematin (Htn) being easily soluble in alkaline solu- 
tions or in acidulated alcohol, shows a different color in both. 
The alkaline solutions are of a reddish hue (Stokes’s red 
hsematin) in thick—green in thin layers ; the acid solutions 
are brown (Stokes’s brown hsematin) in any thickness of 
the layer. 
In each of these solutions, the least absorption takes 
place in the extreme red up to the line B; and both pre- 
sent, if sufficiently diluted, a characteristic absorption-band 
in the red, between C and D. But the very faint, broad 
band of the alkaline solution (the Htn band) is located near 
D (between C and D)—while the narrower and dark band 
of Htn dissolved in alcohol and SO3 is situated close by 
C, between C and D (band of Htn in acid solution, Hoppe). 
This solution also shows frequently three other bands, two 
at the situation of the Og-Hb bands (differing from them as 
to breadth and intensity) and one between b and F (Hcema- 
toin Spectrum, Preyer).. 
Valentin * was the first to distinguish the two bands— 
the first, the alkaline band as the “ Htn-band,”—the other 
(as produced by concentrated acetic acid), the “ Hceviin- 
band.” Preyer calls the first one “ Oxygen-Alkali-Haem- 
atin ” ; and the second “ Acid-band.” 
* Loc. cit. pp. 80 and 82. 26 
THE USE OF THE SPECTROSCOPE. 
Valentin’s Haemin-band, and Preyer’s acid-band have the 
same position, viz.: between C and D, near C, extending 
beyond, and being nearly bisected by this Fraunhofer’s line. 
Though Hoppe’s band of Htn in acid solution has not 
quite the same position, it is nevertheless the same, as the 
situation of these so-called “acid-bands” varies with the 
kind of acid used, and even with the quantity employed. 
Their position can only be generally determined as between 
B and D. The displacement of the band towards A is in 
direct proportion to the quantity'of the acid used (Preyer). 
A Htn solution suitable for Spectrum Analysis may be 
prepared, (according to Valentin, after the most essential 
features of Wittich’s method) in this manner: Blood is 
shaken up with a solution containing one part of carbonate 
of potassium and two of water (in the proportion of one 
part of blood to 6 or 8 of solution), and subsequently fil- 
tered. Of the chocolate-colored precipitate, a portion is 
boiled with aqueous alcohol. 
, A “ Haemin ” solution (Hm) for spectroscopic purposes 
is prepared by heating a blood-solution, to which acid. acet. 
concent, had been added. From this Hm solution a fluid 
presenting the Htn spectrum can be readily prepared by 
adding ammonia to the former. 
Htn (containing O) is also reducible by any of the above- 
quoted reducing agents. After adding any of them to a 
Htn solution, the fluid changes its color, and two bands, 
between D and E, the bands of reduced Htn are presented 
in place of the one. 
They are easily distinguished from the normal bands,— 
the first normal band being very near D, while the first band 
of reduced Htn is so far distant from D, that a considerable 
part of the green-yellow is distinctly observable between it THE USE OF THE SPECTROSCOPE. 
and D. Moreover, this band is made especially prominent 
by its great intensity and darkness. The second, a very 
faint band, includes the line E, and extends to or over b. 
According to Hoppe, the spectrum bands of the reduced 
Htn (Stokes) are identical and belong to the alkaline solu- 
tion of Hcemochroniogen (a primary product of decomposi- 
tion of Hb by acids or alkalies). Hence Hoppe calls Htn 
the product of oxidation of Hsemochromogen. 
The second product in order of importance is the so- 
called Methcemoglobin, an intermediate result of the spon- 
taneous decomposition of Hb into Htn and albuminous 
matter. The band, by which it is characterized, is situated 
between C and D, nearer C. It appears as soon as a de- 
composition of Hb sets in ; and, according to Preyer, even 
when a pure Hb solution is set aside for a couple of hours, 
at the ordinary temperature. Sorby remarks that it is 
found also in the living body, in the dried blood-scabs of 
wounds in the process of healing. Some authors identify 
it with the acid-band. But, by bringing a little acetic acid 
into absolution of Methb before the slit of the spectroscope, 
the difference may be readily perceived (Preyer),-a displace- 
ment of the band towards B taking place instantly. Never- 
theless, it is produced by one of the weakest acids, CO3 
(Lankester). 
We close this chapter with a note on Hccmatoidin and 
Lutein. 
Haematoidin crystals (identical with Bilirubin, according 
to Hoppe) give two bands ; one strong band between b and 
F, very near F, and a fainter one between F and G. 
Lutein (Thudichum,* 1869), the yellow pigment of the 
* Thudichum, Lutein. Centralblatt, 1869. Also Report of the Medical Officer of 
the Privy Council. London, 1870. 28 
THE USE OF THE SPECTROSCOPE. 
corpora lutea of mammal ovaries, and of yolk of egg, absorbs 
in concentrated solution the blue and violet light strongly. 
When diluted before the slit by alcohol or ether, it 
presents two absorption-bands, one including the line F, 
extending more towards G, the other midway between F 
and G. CHAPTER IV. 
PRACTICAL APPLICATION OF THE BLOOD SPECTRUM. 
The Blood spectrum is employed, in the first place, for 
the qualitative demonstration of Hb, for the purpose of 
recognizing blood in secretions (urine), transudations, or in 
pathological products in general. And further, in Forensic 
Medicine, in suspicious stains. 
Secondly, for the quantitative determination of the 
blood-red in the blood of an individual (in physiological 
or pathological cases), and of the quantity of blood in 
general. , 
In order to decide the presence of blood in urine by 
the Spectrum test, the urine has first to be sufficiently 
diluted and filtered. In the fresh urine, within the first 24 
hours, the normal bands will be perceived, if it contain un- 
decomposed blood; if the blood is decomposed, a haematin 
band will appear. The latter will be presented either with- 
out any special treatment of the urine, or it may become 
necessary to previously boil the urine and treat the coagu- 
lum obtained, with acidulated (SO3) alcohol, gently heating 
it. In instances where no blood-corpuscles can be detected 
by the microscope in a blood-colored or ink-black urine (as 
in scurvy, putrid, typhoid, and malignant fevers, or after 
inhalation of AsH.3), the spectrum test will be specially 
important. 
When blood is intermixed with vomited matter, it will 30 
THE USE OF THE SPECTROSCOPE. 
certainly appear undecomposed in the case of profuse hem- 
orrhage from the stomach; but the so-called coffee-ground 
matter contains only haematin. To separate it from foreign 
pigments derived from food or medicine, the fluid must be 
warmed with a little diluted nitric acid, filtered, and the 
residuum dissolved in diluted caustic soda. The faint Htn 
band is then perceived before the slit, but it will be found 
better to change it by a reducing agent, into the two bands 
of reduced Htn. 
The spectral relations of blood, extravasated into the 
gall-bladder, are analogous to those of blood admixed with 
urine. It will there appear unaltered only in the absence 
of bile:' For bile dissolves not only at the temperature of 
the body the blood-corpuscles, but it transforms Hb into 
Htn and albuminous matter. Hb, changed in such a man- 
ner, is occasionally found in the gall-bladder in the form of 
crumbled precipitates, which dissolved in a little diluted 
caustic soda, will give the spectrum of Htn. 
Haematin is a frequent constituent of faeces, the pig- 
ment originating in the normal intestine from animal food, 
but appearing also whenever an hemorrhage takes place in 
any part of the intestine, the lower section of the colon ex- 
cepted ; provided the functions of the intestine be not 
entirely disturbed, as in cholera, where the blood corpuscles 
are found unaltered in the dejections. For the purpose of 
demonstrating Htn in the faeces, an extract has to be made 
with cold alcohol, the residue boiled with alcohol, and a few 
drops,of SO3 added. This extract condensed by evaporation 
is then examined in the spectrum. (Hoppe-Seyler, Hand- 
buch der physiol, und pathol. Chem. Analyse, 4th ed., 1875.) 
Pathological occurrence of Hb or Htn in serous fluids is 
easily demonstrable by the Spectroscope. THE USE OF THE SPECTROSCOPE. 
31 
Tn forensic demonstration,* the object under examina- 
tion (the stain) should be first dissolved in ammonia water. 
This not only liquefies dry blood with greater facility than 
pure water would, but at the same time gives a brighter 
color to the solution, and will make the absorption-bands 
far more distinct, an advantage which will be especially 
appreciated in the examination of very minute quantities 
of the pigment. Of course the solutions must not be turbid. 
If they are so from dust or insoluble albuminous matter, 
precipitated through the spontaneous decomposition of the 
blood, they ought to be filtered. A deficient concentration 
may be compensated by making the observation through a 
thicker layer, or using a greater intensity of the transmitted 
light. 
If we in such a manner succeed in discovering the 03- 
Hb bands, there is scarcely a doubt about the presence of 
blood. However, the reduction-test will have to be made, 
since certain liquids present similar bands. A diluted am- 
moniacal solution of carmine, for instance, yields two bands 
in a similar position, but which maybe readily distinguished 
from the blood-bands, the first being the fainter, and the 
second the darker one, in the carmine spectrum. The posi- 
tion is, besides, not at all identical with that of the blood- 
bands, as will be seen by comparing the spectra 16 and 17 
in the table. The reduction-test, would, of course, have no 
effect upon the carmine spectrum. 
Again, a solution of aniline red gives a single band 
(Table, Fig. 10), similar to the one band of reduced Hb ; it 
will be well, therefore, to carry O to the blood after the re- 
duction, in order to restore the 00-Hb bands, which pro- 
* Hoffman, E., Professor at Insbruck. Einiges ueber forensische Untersuchun- 
gen von Blutspuren. Eulenberg’s Viertelj. 1873. 32 
THE USE OF THE SPECTROSCOPE, 
cedure would not produce any change in the case of the 
aniline solution. 
Other reactions, as Htn or Hm formation, may evidently 
be applied to the diagnosis of blood in forensic cases. 
Concentrated solutions from less recent blood stains will 
frequently present, besides the ordinary bands, the Methae- 
moglobin-band, which in a strong dilution, fades and ulti- 
mately disappears, while the Os-Hb bands still remain dis- 
tinctly visible. Nevertheless, the appearance of this band 
is not applicable to the determination of the age of a blood- 
trace in forensic cases. E. Hoffman (loc. cit. p. 134) has 
shown by experiments, that the existence of Mthb in a 
blood-solution by no means allows the conclusion that the 
blood stain in question is of a remote date ; and that on the 
other hand, the absence of this band does not furnish abso- 
lute proof that the blood under examination is of more 
recent origin. 
The spectroscopic examination of blood stains will be of 
especial value in those instances where the microscopic 
demonstration of the form-elements of blood (particularly 
the corpuscles) is prevented through the action of water 
(water colored bloody by washing hands or clothing covered 
with blood). 
As to the general value of the method, it must be re- 
marked that the non-appearance of the blood-bands by no 
means justifies the conclusion that blood is absent. 
In the first place, the .solution may have been mixed 
with some other pigment by which the blood-bands might 
be obscured. Further, we know by Valentin’s researches,* 
that putrescent old blood may occasionally be met, present- 
* Valentin, G., Histologische und physiologisdie Studien. Ztsclir. f. Biologic, 
vi. 1870, THE USE OF THE SPECTROSCOPE. 
33 
ing no trace of bands, whether diluted by a weak solution 
of caustic, or carbonate of potash, or by a stronger one of 
lodide of Potassium. Proper reagents may, however, even 
in such solutions, produce characteristic bands (sulphide of 
ammonium produces according to Valentin, in such blood 
two reduction-bands instead one,—-the bands of reduced 
Htn?), or even the ordinary bands may be produced by 
introducing nitric oxide gas (NOs), i- e. bands identical with 
the normal bands as to position (see page 47). 
We have so far spoken only of solutions. Now, the 
blood may certainly be brought immediately before the slit 
in thin payers, dried on glass or other transparent substances. 
But in this case one would be hindered in the application 
of reducing agents or other reagents. If, however, the 
material at disposal be very small, so that a solution of 
proper quantity and concentration be not attainable, the 
procedure recommended by Hoppe-Seyler (Handbuch etc.) 
may be followed. The aqueous solution of the stain, freed 
from all contaminations, is dried in a watch-glass protected 
from dust, and with the watch-glass brought directly before 
the slit. On account of the fissures forming after awhile, 
and disturbing the vision, it is advisable to make the 
examination while the stain is yet wet. , 
Since Hb in dried blood is gradually transformed into 
Mthb, which, in a thin layer, will not easily present its 
absorption-band, it may happen that the unaided eye at 
least may not discover any trace of blood bands, if the last 
method be employed (Hoppe). 
For the quantitative determination of the blood-red in 
any blood for physiological or pathological purposes, a Col- 
orimetric method has been applied. It consists in compar- 34 
THE USE OF THE SPECTROSCOPE. 
ing the intensity of the color of the blood, diluted by water, 
with the color of a Hb solution of known concentration. 
This method is pronounced unreliable by Preyer,* who in 
1866 substituted his spectrum method, which is based upon 
the fact that concentrated Hb solutions are, in a certain 
thickness of the layer, impervious to all rays except the 
red (even with strong light), while less concentrated solu- 
tions of the same thickness will leave unabsorbed a part of 
the green, besides red and orange. Thus by diluting with 
water a measured blood-quantum before the slit, until the 
green appears in the spectrum, the percentage of Hb in 
such blood can be determined, provided the amount of Hb 
be known for a test solution, which is (under the same cir- 
cumstances) just pervious for the green. 
The amount of Hb in the blood of the sick, has been 
made the subject of investigation by H, Quincke f (ac- 
cording to Preyer’s method). Quincke has modified that 
method by making use of a hollow glass prism instead of 
the plane-parallel vessel, so as to vary the thickness of the 
layer, the dilution remaining unaltered. 
He has compared the amount of Hb in the blood of 
patients suffering from Chlorosis, Leukaemia, Nephritis, 
Diabetes, Pyaemia, Ileotyphus, etc., with that of normal 
blood, and found, (the latter being assumed as = 1) a very 
considerable diminution in Chlorosis (0.36), Leukaemia lie- 
nalis (0.39), Nephritis (0.58); an increase, on the contrary, 
in a case of Diabetes (1.10)—results, which, of course, have 
only an individual signification.^: 
* Annalen der Chemie und Pharmacie, 1866 ; and “ Die Blutkrystalle,” 1871. 
t Quincke, Ueber den Hasmoglobingehalt in Krankheiten. Virchow’s Archiv. 
54, 1872. 
t Compare also Subbotin, V., Mittheilung ueber den Einfluss der Nahrung auf 
den Hasmoglobingehalt des Blutes. Ztschr, f. Biologic, VII. 1871. And—Coze et 
Feltz, Recherches expgrimentales. Gaz, Med. de Strasbourg, 1869. THE USE OF THE SPECTROSCOPE. 
35 
We cannot omit a reference to Vierordt’s researches, 
who, since 1871, has made the quantitative spectrum de- 
termination of coloring matter in general the subject of 
very elaborate investigations [Berichte der Deutschen 
Chem. Gesellschaft, IV., and: Vierordt, die Anwendung 
des Spectral-Apparates zur Photometric der Absorptions- 
spectren und zur quantitative!! chemischen Analyse, Tu- 
bingen, 1873]. Originally his method was designed for 
measuring the absorption of light by diaphanous colored 
media, in any special region of the spectrum; but the same 
may be employed for determining the amount of pigment 
contained in any solution. Without going into minute 
details, we may mention two of the various modifications 
of his spectrum apparatus. First, the division of the mov- 
able slit plate into two portions, each provided with a grad- 
uated delicate micrometer screw for reading off the exact 
breadth of the slit. Second, a contrivance in the ocular 
of the observing telescope, allowing the obliteration of all 
parts of the spectrum except that which is to be examined. 
The transparent bodies are brought before the upper half 
of the slit. The intensity of the respective region of the 
spectrum has to be compared with that of the direct spec- 
trum incident through the lower slit. This latter is nar- 
rowed gradually until the intensity of both spectra becomes 
equal, and in this manner the ratio of the widths of the 
two slits furnishes a measure for the amount of absorption 
of the medium under examination. 
In conclusion, we have to mention a fact which may 
ultimately be of importance with regard to the examination 
of the blood of the sick ; namely, the possibility of perceiv- 
ing the absorption phenomena of the blood in the living 36 
THE USE OF THE SPECTROSCOPE. 
organism. According to Hoppe the normal bands are 
visible in the ear of man, of white rabbits, the edges of 
the fingers held adjacent, the palm of the hand, and the 
natatory membrane of frogs. 
According to Vierordt [Physiologische Spcctral-Analy- 
sen. Ztschr. f. Biologie, 1875], the fourth and fifth fingers 
are brought into contact, so as to allow the sunlight to be 
visible only through the soft parts; the boundary-line of 
both fingers will then appear of a much brighter red than 
the scarcely translucent phalanges. This boundary-line 
must be laid upon the slit, when the bands will be per- 
ceived and may be determined. Vierordt saw even the 
reduction-band appearing in the spectrum, after putting 
rubber rings around the first phalanges of both fingers, 
compressing, thereby the soft parts sufficiently to cause an 
interruption of the circulation* CHAPTER V. 
ON THE ABSORPTION-BANDS OF BILE AND URINE. 
FRESH ox-bile has a green color, and shows in the spec- 
trum, if the layer be rather thick, a band between D and E, 
nearer D (Hoppe). 
Fel tauri inspissatum, treated with muriatic acid, pre- 
sents this band very distinctly. 
The four best known bile-pigments, Bilirubin, Biliver din, 
Bilifuscin, and Biliprasin, yield (only after treatment by 
nitric or muriatic acid [oxidation]) a pigment, characterized 
by four absorption-bands—the first appearing at C, the 
second before and close to D, the third just behind D, and 
the fourth near to and before E, This pigment (also ob- 
tained from ox-bile, after it has been left standing), has 
been called Choleverdin by Stokvis, and Bilicyanin by 
Heynsius and Campbell.* 
Our more intimate knowledge of the spectrum appear- 
ances of bile and urine dates from Max Jaffe’s researches,f 
1868. He made the interesting discovery that the well 
known Gmelin’s test for bile-pigment (by nitric acid, con- 
taining nitrous acid), which presents a scale of colors (green, 
blue, indigo, red, and yellow), causes characteristic changes 
* Heynsius und Campbell, die Oxydationsprodukte der Gallenfarbstoffe und ihre 
Absorptionsstreifen. Pflueger’s Archiv f. Physiologic, 1871. 
f JaffS, M., Beitrag zur Kenntniss der Gallen—und Harnpigmecte. Centralblatt. 
1868, No. 16. 
Jaffe, M., Zur Lehre von den Eigenschaften und der Abstammung der Harnpig- 
mente. Virchow's Archiv. 47, p. 405. 38 
THE USE OF THE SPECTROSCOPE. 
in the spectrum. As soon as the color of the solution ap- 
proaches the blue modification, a broad dark absorption- 
band appears between C and D, commencing nearer D, and 
extending midway between D and E. On dilution, this 
band separates into two indistinctly circumscribed bands 
(a and /3), with a brighter interval at D. These remain vis- 
ible, though with decreasing intensity, until the appearance 
of the red modification. Almost simultaneously with a and 
/3 (but generally somewhat later), between b and F, almost 
exactly bounded by F, a third band (y) becomes visible, 
which is proportionately more distinct as a and /3 fade away. 
The pigments corresponding to these bands, which ap- 
pear only in the acid solution, have been isolated by Jaffe. 
The reddish filtrate of an extract of human or dog’s 
bile (this extract being made by means of diluted muriatic 
acid) shows in proper dilution the band y much stronger and 
more sharply defined. This solution alkalized by caustic 
soda turns bright yellow, y vanishes, and, after a few min- 
utes, a new band (cl) appears, likewise between b and F, but 
nearer b. It is less distinct, if ammonia be used instead of 
soda. Both bands may appear together, if the alkali has 
been added in slightly larger quantity; a trace of acid or 
alkali would then alternately suppress the one or the other. 
By alkalizing this acid solution, a fourth band is produced. 
The same pigment, giving the three absorption-bands, 
is found preformed, according to Jaffe, in gallstones. 
It is a remarkable fact that the pigment produced by 
diluted muriatic acid from bile is identical with one of the 
normal urine, called by Jaffe Urobilin. Normal human 
urine, without chemical treatment, frequently presents— 
though indistinctly—the band y, when the layer is suffi- 
ciently thick (3-6 Cm), and examined in a strong light. It THE USE OF THE SPECTROSCOPE. 
39 
will be much more distinct in the normal urine, if the latter 
(1-200 Ccm) be precipitated by acetate of lead, the pre- 
cipitate decomposed by diluted sulphuric or oxalic acid, and 
the reddish yellow filtrate examined. Upon adding caustic 
soda, the displacement of the band towards b (the band 6) 
is then also perceived. 
But all these spectrum characteristics may be, with the 
greatest facility, perceived in urine of greater concentration ; 
viz., in the bright red urine of fever patients, examined with- 
out previous preparation in a testrtube. The spectrum ap- 
pearances are not destroyed by boiling the urine with dilu- 
ted mineral acids or alkalies ; even when set aside for weeks, 
and having become completely musty, it will present the 
bands, so long as it has not turned putrid,—which proves 
that the coloring matter of the urine is not so easily de- 
composable as generally thought to be. 
Urobilin is not only distinguished by characteristic spec- 
troscopic qualities, but by brilliant fluorescent phenomena, 
which are produced with facility by treating urine with 
ammonia and ClZn,—Jaffe found further, that fresh pale 
urine, which would not show a trace of an absorption-band, 
will often become darker after remaining exposed to the air, 
and will present the band characteristic of Urobilin. 
This proves (according to Jaffe and Hoppe) that in nor- 
mal urine, a “chromogen” of Urobilin exists; which as 
Jaffe has shown by experiments, is transformed into the 
pigment by taking up oxygen. 
Closely connected with the whole subject is a fact of 
great scientific importance, established by Hoppe [Hoppe, 
Einfache Darstellung von Harnfarbstoff aus Blutfarbstoff. 
Berichte der Deutschen Chem. Gesellschaft, 1874.] At a 
prior date, a reduction-product of Htn, obtained by treating 40 
THE USE OF THE SPECTROSCOPE. 
Htn in alcoholic solution with Tin and CIH, had been de- 
scribed by Hoppe, and supposed by him to be identical with 
Jaffe’s Urobilin ( = Maly’s Hydrobilirubin); but recently he 
found the same pigment to be obtainable with equal facility 
by treating undecomposed Hb in alcoholic solution with Tin 
and CIH. Hence, the coloring matter of the normal faeces 
and urine may be conceived as a product of decomposition 
of the blood-pigment, altered by reduction. 
Bilirubin and Biliverdin also represent intermediate 
stages of this metamorphosis, or they are, at least, closely 
related to blood-pigment. 
Though similar views have long been held, still they 
have only attained a secure basis through Hoppe’s discovery. 
Hoppe remarks, at the close of the paper we have 
quoted : “ Physiological questions of great importance are 
hereby presented. For instance, it will be possible to find 
how great, under normal or pathological relations, the decay 
of blood-pigment or of red blood corpuscles will be in a 
definite time,” 
Pettenkofer’s well known test for bile-acids (by cane- 
sugar and SO3) is liable to disturbance by the presence of 
albumen, since albuminous matter treated by SO3 yields a 
similar purple color. By the use of the spectroscope, the 
absence of albumen may be established. According to 
Schenk (Hoppe, Handbuch, etc., 1875, p. 104), the purple 
solution of bile acids treated by SO3 and sugar, presents in 
proper dilution with alcohol, a band between D and E, next 
to E; and a second band before F, which does not take 
place, if albumen be treated in the same manner.* 
The spectral relations of the bile-pigments are thoroughly discussed in Prof. 
J. C. Dalton, Human Physiology, 6th edition, Philadelphia, 1875. THE USE OF THE SPECTROSCOPE. 
41 
The partial identity of the spectrum (y) of bile-pigment 
and Urobilin (if the latter be considered as a normal color- 
ing matter of the urine), will, it appears, prevent a ready 
application of the spectroscope for discovering bile-pigment 
in icteric urine. 
Stokvis (1870) treated icteric urine by ClZn and am- 
monia in excess, and found three bands : one behind C, 
another at D, and a third between b and F. .It was only 
in icteric urine that he obtained the two first; the third 
was also found in normal urine. 
As a product of decomposition of Indican, Indigo occurs in putrescent 
urine, for the spectroscopic diagnosis of which the absorption ap- 
pearances of Sulphindigotic acid (Indigo dissolved in cone. SOs), 
may be employed. It shows an absorption between C and D, in 
thicker layers extending over D. (See Table, No. 7.) CHAPTER VI. 
ON SPECTRUM ANALYSIS OF THE BLOOD AFTER INTRO- 
DUCTION OF FOREIGN GASES. 
The modifications of the normal blood-spectrum, aftei 
the introduction (or inhalation) of foreign gases, are of 
course based upon the various alterations which the Hb 
thereby undergoes. 
Certain gases expel O from its combination with Hb ; 
and, supplanting it, form new compounds, of greater stabil- 
ity, with Hb; thus preventing it for a longer or shorter 
term, from recombining with O. 
Others have the property of reducing Os-Hb, when 
introduced for a protracted period, even at a low tempera- 
ture, but do not combine with Hb, which will again take 
up O on shaking the solution with air. To this class 
belong Plydrogen and Nitrogen. 
Certain gases have not only a reducing but a decompos- 
ing effect; and, lastly, there are gases by which the Hb will 
be completely destroyed. 
Before discussing these spectrum appearances in detail, 
we have first to make some remarks on the special methods 
adopted in such investigations. 
In general there are three methods employed in these 
researches: 
Ist. By direct introduction of the gases into Hb-oi 
blood solutions. THE USE OF THE SPECTROSCOPE. 
43 
2d. By experimenting with the gases upon animals. 
3d. By observations of cases of gas-poisoning in man. 
This third method has been applied, so far, only in 
reference to the action of two gases, CO and CyH. 
For a number of the other gases, experiments have been 
instituted in accordance with the first and second methods. 
Setting aside the fact that conclusions deduced from 
experiments upon animals can only be accepted with the 
greatest caution as bearing upon man, these experiments 
present so striking a contrast in their results, compared 
with those obtained from direct introduction into blood- 
solutions, that for the time we shall be scarcely tempted 
into drawing such conclusions. 
For instance, we find in Hirt’s “ Gasinhalations-Krank- 
heiten der Arbeiter,” 1873, that, while COg directly intro- 
duced into blood produces the reduction-band, the blood 
of animals killed by inhalation of the same gas only present 
the normal bands, which may subsequently be reduced ad 
libitum. 
The same contrast is found in all comments on spectrum 
appearances as to HS, SO3, and NH3; that is, in reference 
to all those gases in which experiments, after either method, 
have ever yet been made (CO and CyH being here excluded 
from discussion). 
Now we are certainly taught by physiological chemistry 
(vide Kiihne, Lehrbuch der physiol. Chemie, p. 247) that 
“ the alterations in the blood after poisoning are identical 
with those produced by the same poisons in blood freshly 
drawn from the body. This holds good especially for CO2, 
CO, HS, for AsHs, and SbHa.” 
But, remarkable as the contrast pointed out may be, we 
must refrain from concluding, a priori, that it is impossible. 44 
THE USE OF THE SPECTROSCOPE. 
We shall see, on the contrary, that not only have good and 
valid reasons been brought forward for such a difference in 
action by the most eminent men of science, but that an 
attempt to controvert these results would, for the time, 
have but little effect. 
It has formerly been the practice in experimenting upon 
animals, to examine the diluted blood without excluding 
the atmospheric air. Hb having a great affinity for O, will 
combine with it, through mere dilution by water, and will 
thus yield the normal bands, unless made incapable of ab- 
sorbing O, by such a gas as CO. 
The exclusion of air in these examinations was entirely 
unknown before 1867. We are indebted for this method to 
Prof. Gwosdew,* (Krakow) who conceived the idea of ap- 
plying Spectrum Analysis to the forensic diagnosis of 
death from suffocation. The fact of the absence of oxy- 
gen in the blood of suffocated animals had been previously 
made known by Setschenow ; that the air contained in the 
lungs of the suffocated animal was deprived of O, had also 
been demonstrated by W. Muller. Gwosdew consequently 
expected to obtain the reduction-band in the blood of 
suffocated subjects, provided the atmospheric air could be 
completely excluded during the examination. By means 
of a certain procedure (described loc. cit.) he indeed suc- 
ceeded, in several experiments, in obtaining the postulated 
band ; whence he concluded death by suffocation to be 
diagnosticable by that means. The experiments were re- 
peated by F. Falk,* (Berlin), with an apparatus of his own 
invention, with entirely negative results, due, as was shown 
* Gwosdew, J., Spectroscopische Untersudumg des Elutes bei Erstickten. Rei- 
chert’s Archiv. 1867. 
* Falk, F., Verwendung des Spectroscops zu forensischen Zwecken. Pragei 
Vierteljahrschr. 1869. THE USE OF THE SPECTROSCOPE. 
45 
by Kotelewski,* (Warsaw) in 1870, to some defect in its 
construction, K. at the same time relating the positive 
results he had obtained by Gwosdew’s method as well as 
his own. 
A test-tube, narrowed and drawn out at the 
upper end, is filled (after 'warming it,) with pure 
glycerine, mixed with a little distilled water, to 
the level of b. Upon the open end a short rub- 
ber tube is fixed, and tightly bound with thread ; 
the fluid is then subjected to a boiling heat, in 
order to expel the air from the mixture. The 
rubber tube is then drawn over the thick end of a 
metal tube, provided with a stop-cock, and at the 
opposite end with a pointed canula, the bore of 
the latter being somewhat wider than that of a 
common Pravaz syringe. The rubber tube is now 
bound over tightly with thread ; the boiling of the 
glycerine continued cautiously for a little while, 
until the level of the fluid reaches about bl; the 
canula-point is now immersed in a vessel filled 
with thoroughly boiled glycerine, and the stop- 
cock closed. 
Then the glycerine, completely deprived of 
air, is allowed to cool. 
After cooling, the apparatus is ready for the 
reception of the blood, the point being inserted 
His apparatus and method of procedure are as follows : 
in the heart or one of the larger veins; and, as 
Fig. 4 
soon as the stop-cock is opened, the blood rushes into 
the vacuum. The stop-cock is now closed, the apparatus 
shaken, and brought before the slit. 
* Kotelewski, Zur Spectral-Analyse des Blutes. Centralblatt, 1870. No. 53 und 54. 46 
THE USE OF THE SPECTROSCOPE. 
Although obtaining the reduction-band in all his experi- 
ments, Kotelewski came to the conclusion, nevertheless, 
that Spectrum Analysis was not applicable to the diagnosis 
of death from suffocation, for the following reason: He had 
found the blood of every dead animal or human subject 
under examination (atmospheric air being excluded) to 
present the absorption-band of reduced Hb, and that it was 
absolutely impossible to obtain the Oa-Hb bands under 
those conditions. The tissues, he says, consume oxygen so 
rapidly as to leave only reduced Hb in the blood of the 
veins a few minutes after the lungs have ceased carrying O 
to the blood. He has not omitted to examine the blood 
from a vein of a living animal in the same way, and found 
it to contain Og-Hb. All these remarkable observations of 
Kotelewski were subsequently confirmed by Falk, in a short 
paper published in the “Deutsche Klinik,” 1872. It is 
therefore an established fact that Spectrum Analysis is 
decidedly inapplicable to the medico-legal diagnosis of 
death from mechanical suffocation. The subject is treated 
in this sense in Caspar’s “Hand-book of Legal Medicine,” 
edit. 1871. 
There is no doubt but that such a mode of investigation 
in gas-poisoning will, in certain instances, give quite a dif- 
ferent result from what has been formerly taught; though 
it will be apparent that the practical effect,'as to a' post- 
mortem diagnosis, will be as nugatory in poisoning by 
“reducing” gases, as it has been shown to be in mechanical 
suffocations. Nevertheless, it may become a very useful 
means of inquiry in experintents upon animals, as well as in 
observations upon man, if undertaken before death has set 
in. On the other hand, two bands equal in position to the 
normal bands, would, if obtained from a dead body (the THE USE OF THE SPECTROSCOPE. 
47 
blood being examined under exclusion of air,) certainly go 
far to prove a deep alteration of the Hb, especially if 
found irreducible. The reduction-test could not in such 
cases be made in the usual manner, unless the reagent 
could be admitted without the introduction of air. 
Kotelewski’s method might probably be supplanted by 
one which could be performed with greater, facility; viz., 
the examination of the undiluted blood, in a thin layer, 
taken from the animal in an exhausted vessel or glass-tube. 
We shall see, further on, how other incidents beside the 
admission of air, may possibly prove the cause of the strik- 
ing incongruity we have pointed out. 
We now proceed to discuss in detail the spectrum ap- 
pearances of the blood with regard to various gases, or 
vapors, though some, such as bisulphide of carbon vapors, 
exhibit no influence whatever upon the normal spectrum. 
The so-called indifferent gases, Hydrogen and Nitrogen, 
we have already alluded to. 
Three gases, N02, CO, and CyH, are known to expel 
oxygen from its combination with Hb, supplanting the 
oxygen, and forming stronger compounds with Hb. 
N02. Nitric oxide (binoxide of nitrogen) is prepared 
by treating copper clippings with nitric acid. (Deoxida- 
tion of NOS by the metal.) 
After introduction of nitric oxide gas into 02-Hb solu- 
tions, the normal bands appear unchanged. They no 
longer pertain to Q2-Hb, but to a newly-formed compound, 
discovered by Prof. L. Hermann,* and prepared by him 
in crystal-form, N02-Hb. Hermann introduced N02 (by 
* Hermann, L., Ueber die Wirkungen des Stickstoffoxydgases auf das Blut 
Reichert’s Archiv. 1865. 48 
THE USE OF THE SPECTROSCOPE. 
exclusion of air) into a blood-solution, previously deprived 
of O by Hydrogen gas. The blood became of a bright 
crimson color, and presented, instead of the reduction- 
band, two bands equal in position to the 02-Hb bands, but 
of a less intensity. These bands cannot be transformed by 
the addition of a reducing agent. 
Since it can never be inhaled, being converted by the 
contact with air into the irrespirable Hyponitric acid (NO4), 
the knowledge of its spectrum relations to blood has ap- 
parently no practical value. It is, however, spectroscopi- 
cally interesting, in that it shows that hands almost entirely 
identical with the normal bands may have quite a different 
signification. It may also be used, as mentioned above, as 
a means of detecting the presence of blood in cases where, 
from its age, the original normal bands are entirely absent. 
CO. This gas is readily prepared by heating oxalic acid 
in a flask with five or six times as much cone. SO3. The 
acid being resolved into water, CO2 and CO, the gases must 
be passed through several wash-bottles filled with a solution 
of caustic soda (half dilution) for the absorption of C02. 
Pure carbonic oxide gas, introduced into 02-Hb solu- 
tions, expels oxygen, supplanting it, and forming the crystal- 
Hzable compound CO-Hb (Hoppe), known before N02-Hb. 
The bright cherry-red solution yields, instead of the normal 
bands, two bands almost identical to these, which may be 
distinguished by their displacement towards the violet end 
of the spectrum.* The difference of position being scarcely 
sufficient for their recognition, the reduction-test must here 
again decide their character. 
Upon adding reducing agents (Tin solution in prefer- 
* In observations of this kind it is important to know, that by widening the slit 
a perceptible displacement of bands is effected. THE USE OF THE SPECTROSCOPE. 
49 
ence) the CO-Hb bands do not disappear, nor are they 
replaced by a reduction-band—at least, not instantly,—some- 
times, especially at a low temperature, not before a'lapse of 
several days. This remarkable condition exists also in the 
blood of animals killed by inhalation of the gas. 
CO-Hb not being an inseparable compound, though of 
greater stability than Og-Hb, it follows that, as Hoppe sug- 
gested, the examination must be made as early as possible 
after the poisoning has been discovered. With man, how- 
ever, poisoning from pure CO will rarely occur; and it is 
evident that in those frequent cases of poisoning by CO- 
mixtures, such as charcoal fumes* or common gas, the 
spectroscopic appearances will be, from the presence of a 
large percentage of air, somewhat more complicated. 
The bands presented by such blood, which does not 
always show the florid arterial color of CO blood, being 
frequently dark, may be displaced towards the more re- 
frangible end of the spectrum ; but on adding a reducing 
agent, a reduction-band will in such cases appear, due to 
the action upon the oxygen introduced with the fumes. 
That is, the space between the two bands will be obscured 
—but a more minute examination will show that the two 
bands have not disappeared. This, together with the 
changed situation (which may be verified by the simultane- 
ous observation of a normal blood-solution—by means of 
the prism of comparison) will decide the character of the 
blood under examination. 
This is not a mere theory, but the result of an observa- 
tion f made by Prof. Hensen (Kiel) in a case of poisoning 
* A valuable paper by A. Gamgee “ On poisoning by carbonic oxide gas, and by 
charcoal fumes,” is contained in the Journal of Anatomy and Physiology (London) 
I. 1867. 
f As far as the author knows, the only detailed observation on record. 50 
THE USE OF THE SPECTROSCOPE. 
by charcoal fumes, (cured by transfusion,) communicated 
by Jiirgensen. (Drei Falle von Transfusion des Blutes, Berl. 
Klin. Wochenschrift, 1871, p. 304.) 
The diagrams will illustrate Hensen’s observation. 
The spectra 1,2,3 are drawn according to Hoppe-Seyler 
“ “ 4, 5, 6 and 7 according to Hensen 
4. 02-Hb bands (81-84 ; 90-96) 
5. Reduction of 4 (83-94) 
6. Bands of blood from man poisoned 
by charcoal fumes (82-86 ; 91-98) 
7. Reduction of 6 (82-97), produced 
by adding to 6 a drop of sulphide 
of ammonium. 
1. 02-Hb bands (82-88 ; 96-105) 
2. Reduction of 1. 
3. CO-Hb bands, showing the dis- 
placement towards b (83-91 ; 
97-107) THE USE OF THE SPECTROSCOPE. 
51 
Though Hensen has made his observation without de- 
termining the Fraunhofer’s lines, we are, from his descrip- 
tion of the bands, enabled at least to locate his D at 80, 
and his E at a fraction beyond 96. 
The bright interval between the two bands is obliterated 
(7) by a reduction-band (which would not take place in a 
pure CO-Hb solution); but the two bands are not destroyed. 
The contraction by one degree of the scale, observed at the 
right end of 7 was due, according to Hensen, to some 02- 
Hb, which had contributed in his case to extend the sec- 
ond band (spectrum 6) towards b, and which had been 
reduced by the reagent. 
CyH. Cyanide of hydrogen (hydrocyanic or prussic 
acid) also forms a strong and crystallizable compound with 
Hb (Hoppe, Preyer). The spectrum of a solution contain- 
ing CyH-Hb does not present—according to Hoppe—any 
change of aspect from that of a pure 02-Hb solution. On 
adding a reducing-agent, the two bands disappear and are 
replaced by the reduction-band. Nevertheless, this com- 
pound is more fixed than 02-Hb. If, according to Hoppe, 
a blood-solution, to which a few drops of prussic acid have 
been added, be enclosed in a glass tube, so as to leave but 
very little air above the liquid, the mixture will present the 
normal bands, even after the lapse of several months, while 
a few days would suffice for the production of the reduction- 
band, if CyH had not been added. 
Though Preyer’s results differ considerably from Hoppe’s, 
still observers agree in the fact that the gas cannot be spec- 
troscopically demonstrated in the blood of poisoned animals. 
The same negative result appears in the single observation 
recorded concerning the spectrum examination of the blood 52 
THE USE OF THE SPECTROSCOPE. 
of man after poisoning by prussic acid. H. Siegel (Leipzig) 
—“ Fall von Vergiftung durch Blausaure,” Archiv der Heil- 
kunde, 9, 1868—found in this case, the blood dark, wholly 
fluid ; in thin layers it appeared cherry-red, and the two 
absorption-bands of Os-Hb presented themselves in all 
distinctness. 
COg is prepared by the decomposition of marble with 
diluted muriatic acid; or in greater purity, by treating 
chalk with concentrated sulphuric acid. 
After the direct introduction of carbonic acid gas into 
a blood-solution, the normal bands disappear, giving way 
to the reduction-band. On shaking the solution with air, 
it again absorbs oxygen, the Cb-Hb bands re-appearing. 
Nothing of this kind is seen in the blood of animals 
poisoned by COg. As we have previously remarked, only 
the normal bands are perceived, and these may be after- 
wards reduced by any reducing agent. An examination 
with the exclusion of air would doubtless give a different 
result. The questionable value of a reduction-band ob- 
tained under such conditions will be sufficiently demon- 
strated in the introductory remarks to this chapter. 
Although usually nothing is mentioned about COg but 
its reducing action, we should observe that it is not thus 
limited. On leaving a blood-solution into which C02 had 
been introduced, for several hours at a low temperature, 
without shaking it, the so-called Methaemoglobin band will 
appear. The same phenomenon will probably be perceived 
in the blood of animals poisoned by COg. 
HS. After the direct introduction of sulphuretted 
hydrogen gas into a diluted, neutral blood-solution, the THE USE OF THE SPECTROSCOPE. 
53 
liquid turns to deep red ; and besides the normal bands 
which soon become fainter, and disappear after a short time, 
being replaced by the reduction-band, the band appears, 
which is generally considered as characteristic and specific 
for HS blood (Hoppe-Seyler), and is therefore called the 
HS band. It is a well-defined absorption-band situated 
in the orange between C and D, about midway between 
both lines. 
The reduction-band spoken of is, by shaking the solu- 
tion with air, re-converted into the normal bands. 
The chemical character of the pigment, represented by 
the HS band, is not yet thoroughly understood. The gas 
is oxidized in the blood to S and HO, and this is ac- 
complished far more rapidly than it would be by the 
oxygen of the air, in the absence of blood ; whence the 
conclusion that the ozonizing faculty of the blood pigment 
is performing an essential part in this process. Before the 
reduction of Os-Hb, implied in this operation, has been 
completed, the pigment is formed which optically manifests 
itself by the HS band. 
Hoppe concludes from its origin, that the body repre- 
senting it is a sulphur compound, either of Htn or of Hb. 
If the introduction of the gas is continued for a length- 
ened period, even this pigment will undergo decomposition, 
with the separation of sulphur, and an albuminous matter. 
At this stage of decomposition there are, however, no 
absorption-bands whatever visible. 
That the HS band is not to be considered identical 
with the Htn band, as asserted by several authors, Hoppe 
has demonstrated in his paper*: “ Ueber die Einwirkung 
des HS auf den Blutfarbstoff.” The difference is not only 
* Medic. Chem. Untersuchungen. Berlin, 1866. 54 
THE USE OF THE SPECTROSCOPE. 
in the position of this band and those produced from de- 
composition by acids or alkalies, but in the decisive point 
of the reduction-test. The non-identity of the HS and 
the Htn band is proved by Hoppe through the immuta- 
bility of the HS band by reducing agents. While all Htn 
solutions, on adding sulphide of ammonium, soon exhibit 
their two reduction-bands in the green (the band in the red 
disappearing), the HS band will remain unchanged. The 
HS band differs also as to position from the Methzemo- 
giobin band. It is besides remarkable that the latter will 
disappear after gently heating the fluid, while the HS 
band requires the application of boiling heat for this 
purpose (Preyer). 
With a greater dispersion than that used for the 
spectrum table, the respective bands present the following 
positions : 
As to the two bands between D and E, which often 
remain visible in the spectrum beside the HS band, they 
are considered and designated by almost all authors as the 
normal bands. Only Preyer (“ Blutkrystalle,” 1871) 
entertains the view—supported by arguments—that they 
do not pertain to Os-Hb, but—as he says—probably to a 
combination of Og-Hb with sulphur. 
Fig. 6. THE USE OF THE SPECTROSCOPE. 
55 
Now the HS band would certainly offer an excellent 
means of diagnosis in cases of poisoning by this gas. But 
here we have the oft-repeated contrast appearing strik- 
ingly. 
Spectrum observations concerning the blood of man 
poisoned by this gas have as yet not been recorded. We. 
are therefore limited to its results in experiments made 
upon animals. These records, however, afford us no evi- 
dence of the characteristics of this poisoning in the case 
of warm-blooded animals. Rosenthal and Kauffman,* and 
Hoppe have observed “ the higher degrees of blood-decom- 
position ” in cold-blooded animals, it is true; but concern- 
ing that class which more immediately interests us, they 
say “The normal bands are perceived ; but a smaller quan- 
tity of sulphide of ammonium is required for withdrawing 
all O from the blood.” At another place they say : “It 
seems quite self-evident that the primary effect of the gas 
introduced into the circulation must consist in withdrawing 
the O. That the other effects observed .by us, the decom- 
position of Hb and Htn do not appear, is easily explainable, 
since death must set in, as soon as the larger proportion of 
O is withdrawn from the blood.” 
L. Hermann f is very explicit on the point in question. 
He says: “It is certain that, with warm-blooded animals, 
only the first stage of the described effects upon the blood' 
can be brought about, as the reduction of the blood alone 
must, by necessity, cause immediate death from suffocation. 
There is, indeed, nothing to be seen in the blood of poisoned 
warm-blooded animals, of the absorption-band of the Hsema 
tin-like pigment above described.” 
* Rosenthal, J., und Kauffman, “ Ueber die Wirkungen des HS Gases auf den 
thierischen Organismus.” Reichert’s Archiv. 1865. 
t Hermann, L., Lehrbuch der experimentellen Toxicologic. Berlin, 1874. 56 
THE USE OF THE SPECTROSCOPE. 
Eulenberg, * who has made a number of experiments 
upon animals, makes no allusion to the point. Hirt (loc. 
cit. p. 47) follows Rosenthal and Kauffmann. 
In all these spectrum observations, air had been admit- 
ted. In the only observation existing, made with ex- 
clusion of air, by Falk (communicated in the paper quoted 
above, Deutsche Klinik, 1872), the HS band is not al- 
luded to. 
But in Falk’s observation we have another point worthy 
of special notice. He has obtained in his experiment the 
00-Hb bands, whence he concludes that animals poisoned 
by HS have still O in their blood, and that they are not 
therefore suffocated. To us this result appears in the light 
of Kotelewski’s proposition, to offer almost a confirmation 
of Preyer’s view, that the two bands do not appertain to 
unaltered 02-Hb. 
Thus we perceive that the tw’o bands in the green do 
not at present offer anything characteristic for the HS 
spectrum ; and the band in the red not having been found 
(before the slit) in the blood of warm-blooded animals, the 
contrast pointed out appears to be real also in this case. 
But there yet remain some points of qualification. The 
HS band will, under certain circumstances, fail to appear 
even after direct introduction. Thus it is not seen, when 
the gas has been sparingly introduced, or in very diluted 
solutions. The latter peculiarity it has in common with 
other bands in the red. Furthermore, Valentin has, since 
1863, repeatedly insisted upon the necessity of employing a 
bisulphide of carbon prism for bands of this description, 
especially for the HS band. Again HS being partially 
* Eulenberg, H., die Lehre von den schadlichen und giftigen Gasen. Braun- 
schweig, 1865. THE USE OF THE SPECTROSCOPE. 
57 
eliminated through the lungs during the experiment, a 
difference in the spectroscopic results seems possible, ac- 
cording to the longer or shorter duration allotted to the 
procedure of poisoning. With this idea in view, the author 
has made a number of experiments upon rabbits, and has 
obtained the HS band with all desirable distinctness in two 
experiments, in which the animals were killed by the inhala- 
tion of the gas with the rapidity of lightning. 
In seven experiments of a somewhat more protracted 
duration, nothing but the normal bands were perceived. 
In the first two cases the animals were placed under a 
narrow conical glass bell into which the gas had been freely 
introduced before ; in the latter a large roomy vessel was 
substituted. The gas was developed in the usual manner 
by treating SFe with diluted S08, and conducted through 
a wash-bottle, before reaching the bell. The blood was 
taken without excluding the air, and diluted by water. 
The prism employed was Eaton’s compound CS9 prism.* 
the apparatus being adjusted to the following scale : 
F b EDGE a A 
28.8 55-56 63 100 127 136 144 154 
The HS band seen immediately after the death of the 
animal, had the position 118—ill, perfectly identical with 
the band obtained after direct introduction of the gas into 
a blood-solution, or after diluting blood with HS water, 
or after adding sulphide of ammonium in excess to a blood- 
solution. The two normal bands, appearing beside the HS 
band were found reducible. 
It thus appears that, under certain conditions, the ex 
* The band has been seen also by means of a simple flint prism. 58 
THE USE OF THE SPECTROSCOPE. 
periment upon animals with HS will give the same result, 
as the direct introduction into recently drawn blood. 
That this observation, even if confirmed, will not be 
practically advantageous in forensic diagnosis, is evident. 
For, in the first place, the pure HS gas is probably never 
the cause of fatal accidents. Besides, we should, after the 
poisoning of man by HS mixtures, have reason to expect 
the characteristic band only under the conditions above 
described. 
The value and importance of Spectrum Analysis does 
not depend upon its applicability to diagnosis in forensic 
medicine. The reader needs only to be reminded that 
HS, for instance, is employed even as a remedy, which 
fact is certainly sufficient reason why the physician should 
become acquainted with its spectrum deportment towards 
blood. 
In connection with this subject we have to point out a 
possible source of error. In the very locality occupied by 
the HS band, there will appear, if sunlight be employed, 
under certain circumstances, a dark line belonging to the 
so-called terrestrial or atmospheric lines. This line will be 
widened by any blood-solution brought before the slit, and 
may thus deceive us. It is therefore absolutely necessary 
to make observations of this description not exclusively 
with solar light—but at the same time with artificial light— 
in which latter case no absorption lines will be shown. 
S02. Sulphurous acid, for technical purposes, is usually 
prepared by the combustion of sulphur in an enclosed 
place, with a regulated admission of atmospheric air; and 
is obtained pure for physiological experiments by heating THE USE OF THE SPECTROSCOPE. 
59 
SOs with copper clippings (deoxidation). Hirt gives (loc. 
cit. p. 71) the results of his valuable experiments made 
with this gas upon rabbits, dogs, and frogs. According to 
Hirt’s researches the spectrum appearances are; After 
direct introduction of sulphurous acid into a blood-solu- 
tion, the Og-Hb bands disappear instantly, no other band 
appearing in their stead. If the air introduced into the 
blood-solution contained 30 percent. SOg, the bands vanish 
after half a minute; if 10 per cent., after i| minutes. The 
fluid at the same time assumes a peculiar brownish color, 
the Htn band, however, not being perceptible. 
The blood of the animals killed by the inhalation of 
the gas did not present any perceptible change from the 
normal aspect. 
N04 vapors. Fresh ox’s* blood treated by hyponitric 
acid vapors (Eulenberg, loc. cit. p. 247) coagulated rapidly 
to a chocolate-colored mass. 
Gamgee * obtained an acid-band, after treating blood 
with NO3 salts, the Og-Hb bands simultaneously becoming 
very faint. After adding a few drops of NH3, the blood 
was again restored from dirty-brown to its original red 
color, the new absorption-band disappearing, and the bands 
between D and E becoming much more distinct. The 
blood of animals poisoned by NO4 (Eulenberg) has not been 
submitted to the spectrum test. 
NH3. Pure Ammonia is prepared by heating a mix- 
ture of Chloride of Ammonium and moist hydrate of lime 
in a distilling apparatus. 
Koschlakoff and S. Bogomolofif, * in their experiments, 
* Gamgee, A., Note on the Action of Nitric Oxide, Nitrous Acid, and Nitrites 
on Haemoglobin. [Proceedings of the Royal Society of Edinburgh, VI. 1867,] THE USE OF THE SPECTROSCOPE. 
conveyed the gas through a rubber tube cooled by ice. The 
test-tube containing the blood-solution was cooled in the 
same manner. 
Hirt (loc. cit. p. 93) also furnishes the results of his 
experiments. 
Koschlakoff and B. remark that the solutions show 
primarily a yellowish tint, becoming gradually yellowish- 
brown,, and at last brownish-green. 
Hirt found the blood of animals killed by inhalation of 
a mixture of 90 per cent. NH5 and 10 per cent, atmospheric 
air at first a dark red of a bluish tinge, rapidly brightening 
on contact with air. 
All agree as to the effect upon the spectrum on direct 
introduction. The Og-Hb bands gradually disappear, no 
reduction nor Htn band taking their place. 
According to Hirt, the bands reappear after shaking the 
solution for a short time, vanishing after renewed introduc- 
tion of the gas. 
No abnormal condition, however, is perceived in the 
spectrum of the blood from animals killed by inhalation 
of the gas \Hirt\. 
AsHg and SbHg. Koschlakoff and B. state that the 
introduction of arsenetted and antimonetted hydrogen also 
produces a rapid change of color. The solutions change 
to a yellowish-brown, afterwards to a greenish-brown. The 
Og-Hb bands gradually disappear, and are substituted by 
the reduction-band, while the solution becomes slightly red. 
On the following day, the reduction-band vanishes in its 
turn, and traces of the normal bands are visible. The 
* Koschlakoff und Bogomoloff. Ueber die Wirkung des Ammoniaks, des Arsen 
—und Antimonwasserstoffs auf die Blutpigmente. [Centralblatt, 1868. No 39.] THE USE OF THE SPECTROSCOPE. 
61 
general assumption that these gases (especially AsH3) de- 
stroy the 02-Hb bands without reduction, must therefore be 
qualified; such a destruction taking place only in the thin 
layers. 
PH3. Phosphoretted hydrogen has, according to Kosch- 
lakoff and 8,, the same effect as NH3 upon lt 
destroys the bands without reduction or Htn formation. 
Dybkowsky,* however, obtained the reduction-band after 
conveying the gas through defibrinated dog’s blood. On 
shaking the solution with air, the blood changed from the 
dark-brown color it had assumed to a somewhat lighter 
shade, when the normal bands re-appeared. 
The blood of animals poisoned by AsH3, SbH3, and 
PH3 has not been examined spectroscopically. 
* Dybkowsky, W., Beitrag zur Theorie der Phosphor-Vergiftung. (Hoppe-Sey- 
ler’s Med. Chem. Untersuchungen, I. 1866.) APPENDIX. 
NOTE ON THE EXAMINATION OF NARCOTIC POISONS 
AND REMEDIES. 
VALENTIN * has examined with the spectroscope a great 
number of tinctures from narcotic remedies. None of them 
yields a spectrum sufficiently specific to enable us to dis- 
tinguish them from each other. 
He has further submitted to the spectrum test the 
bright-colored liquids, obtained by treating various poison- 
ous alkaloids by sulphuric acid. To a small quantity of the 
alkaloid (Morphine, Narcotine, Strychnine, Brucin and Vera- 
trine), be added from one to two cubic centimetres of strong 
sulphuric acid, containing NOS ; besides which he added a 
trace of NOS and small pieces of crystallized manganese. 
The vivid colors obtained by this procedure at ordinary 
temperature or after boiling, remained unaltered for three 
or four days, frequently for a longer period. 
These researches were, however, equally negative in 
their result; in so far that in no instance were absorption- 
bands met with, which would be the only means of discrim- 
inating between the poisons. 
Researches on the spectral influence of the admixture 
* Valentin, Gebrauch d. Sp., etc., pp. 102 and 105. APPENDIX. 
63 
of alkaloids to blood have been made within the last three 
years by Binz, and by Rossbach.* 
Rossbach found that a blood-solution containing an 
a|kaloid was equally well reduced by sulphide of ammo- 
nium as pure blood. 
Nevertheless, it is a fact that the reduction of blood is 
delayed by alkaloids. Simple heating at from 40 to 50° C 
will reduce pure blood. Rossbach compared with that the 
effect of the admixture of an alkaloid; and found that, in 
this case, a higher temperature was required than with 
pure blood. 
Binz has established the same retardation as to the 
reduction of 00-Hb, in reference to Quinine. 
Both authors connect the effect of the alkaloid with the 
transmission of ozone. 
Rossbach infers that the alkaloids cause the ozone to 
combine more strongly with Hb, so that the latter will not 
be able to transmit ozone to other bodies so easily. 
NOTE ON THE USE OF THE SPECTROSCOPE IN 
OPHTHALMOLOGY. 
It is well known that the prism is largely used in 
Ophthalmology, for its deviation effects; but as our argu- 
ment is confined to the dispersion of light, we shall notice 
only the application of the Spectroscope to cases of Color- 
blindness. 
This affection has latterly engaged the attention of 
oculists in a much greater degree, on account of the prac- 
tical importance it has assumed in reference to many 
* Rossbach. Ueber die Einwirkung der Alcaloide auf die organischen Substrate 
des Thierkdrpers. Verhandlungen der Physic. Med. Gesellschaft in Wurzburg, 111. 
4, 1872 ; and VI. 3 und 4, 1874. 64 
THE USE OF THE SPECTROSCOPE. 
industrial occupations (such as railroad and steamboat 
employes, etc.). 
An individual may be examined as to his reaction upon 
mixed or pure colors,* the former being those shown by an 
ordinary colored material, pigments, papers, etc., reflecting 
various colors, which together form the fundamental color; 
or, by colored glass, liquids, etc., which will transmit not 
one but several colors. 
If, in a case of partial color-blindness, among a number 
of pigments, one is confounded with the others, this will 
have to be examined by prism or spectroscope. 
In order to find the composition of an opaque pigment, 
we look through a prism at a thin stripe of the material, 
laid upon a dark ground (preferably upon black velvet). 
The light, emanating from the pigment, is thus decomposed 
into its several constituents. 
If the object is transparent, it has to be brought between 
the sunlight and the prism, or between the prism and the 
eye, in order to determine the colors so transmitted. 
Pure colors are only such as appear in a spectrum from 
a minimal slit; hence, the reaction of an individual for pure 
colors must be examined by means of the spectrum, which 
has either to be thrown upon a screen by means of a slit 
and prism, or directly upon the retina with the spectroscope, 
the latter method being better and far less tedious. 
The points to be investigated at the examination are: 
whether the individual sees the whole spectrum, as far as a 
normal eye is, able to perceive it, or whether it appears to 
him shorter at one end. Which part of the spectrum 
appears brightest or darkest. Whether he sees a certain 
* “ Chromatoptometry ” : The examination of the eye as to its sensibility for 
rays of light, of various wave-lengths. APPENDIX. 
65 
portion of the spectrum dark (thus manifesting the entire 
loss of perception of a given color). He is then requested 
to relate the succession of the colors; or the parts of the 
spectrum which appear to the normal eye as the fundamen- 
tal colors are isolated. This is done by pushing a black 
screen over the spectrum, with a slit so narrow as only to 
permit the color in question to be seen through it. 
In all instances the situation of the color has to be in- 
dicated, in its relation to the Fraunhofer’s lines. 
We have only to add that the spectrum colors can be 
mixed for the purposes of this examination.* 
Quite recently, Dr. J. Stillingf has introduced in the 
examination of the eye, in cases of partial color-blindness, 
the employment of the bright metal lines. In a number 
of such cases, the bright lines of Lithium, Sodium, Calcium, 
and Thallium, have been used by him, the individual under 
examination indicating the color of the line presented 
through the spectroscope. 
We have previously mentioned colored glass as a means 
of examining the reaction upon colors. Colored glass, when 
employed for spectacles, ought to be examined with the 
spectroscope in reference to the colors they transmit. 
Neither the apparent color-tint, the brightness of the 
transmitted light, nor the thickness, for instance, of a blue 
glass, permit the anticipation as to which particular colors 
will be absorbed or transmitted ; the spectroscope alone 
can determine the matter. 
* Snellen und Landolt, Ophthalmometrologie (Graefe und Saemisch, Hand- 
buch der Augenheilkunde, 111. x, 1874). Maxwell (Philosophical Magazine, 1861). 
Experiments on color. 1855. Helmholtz, Physiologische Optik, 1867. 
t Stilling, J., Beitrage zur Lehre von den Farbenempfindungen. Stuttgart, 1875. 
(Beilage zu den Klin. Monatsblattern f. Augenheilkunde.) 66 
THE USE OF THE SPECTROSCOPE. 
CONCLUDING REMARKS. 
SPECTRUM Analysis must not, for a moment, be con- 
sidered as a modern Philosopher’s Stone, though it has 
already rendered considerable service in the hands of phys- 
iological chemists; and will doubtless lead us to still more 
valuable results. 
We are especially indebted to it for the progress in our 
acquaintance with the coloring matters of the animal body, 
the complicated composition and mutability of which does 
not admit of ordinary chemical analysis. 
It should, however, not be forgotten that the absorption 
phenomena give (properly speaking) nothing but a detailed 
description of color ; and that they, therefore, can only be 
considered and employed as an aid to chemical analysis, 
their chief value consisting in indicating the direction in 
which the chemist must pursue his researches. 
On the other hand, these phenomena will, without 
doubt, enable the physician to diagnose pathological al- 
terations occurring in blood or other animal fluids, wherever 
the absorptive characters of the respective pigments are 
once definitely established. And if Spectrum Analysis be 
not able to solve every forensic problem, it will still be 
useful in toxicology and in experimental pathology, where 
the causes are not sought for, to demonstrate the mode of 
action and effect. 
We have, in the preceding pages, endeavored to furnish APPENDIX. 
the student, and more especially the practical physicians, 
(who are already preoccupied with other and perhaps more 
important branches of study) with a compilation of the 
more salient points of medical Spectrum Analysis, without 
pretending a thorough exhaustion of the subject. 
New York, December, 1875, 68 
THE USE OF THE SPECTROSCOPE. 
SUPPLEMENTARY NOTE TO CHAPTER VI. 
While this treatise was in press, A. Schmidt’s Essay, 
“ Ueber die Dissociation des 02-Hb im lebenden Organis- 
mus,” Jena, 1876, was received, where PreY- 
Eß’s method of examining the blood with ex- 
clusion of air is described, by which, among 
other results, the HS band was observed in 
the blood of warm-blooded animals poisoned 
HS. 
preyer’s apparatus. 
A Pravaz syringe, into which a glass cylin- 
der G has been fitted, is filled with rock-oil 
(naphtha, Steinoel), which does not contain 
O, and does not deprive the blood of its O 
for many hours. Without allowing air to 
enter the tube, the piston is then put into 
place, and the superfluous oil thrown out 
through the canula. 
The glass cylinder must have the form 
indicated by G and g. It must be cut ob- 
liquely at the lower end and plane at one 
side. 
The thin stratum of blood between the 
glass wall of the syringe and the plane side of 
the cylinder serves as object for the micro- 
spectroscope. 
Fig. 7. 
As soon as a sufficient quantity of blood 
from the heart of the animal has been drawn into the 
syringe, the point of the canula has to be closed by 
immersion in warm wax.