A CONTRIBUTION 
TO THE 
ILEMATINIC PROPERTIES OF DIALYZED IRON. 
RUING EXTRACTS FROM COMMUNICATIONS RE Ml) BEFORE THE BOSTON 
SOCIETY OF THE MEDICAL SCIENCES AND h'HE BOSTON SOCIETY 
FOR MEDICAL OBSERVATION. 
BY 
ROBERT AMORYt M. D., 
OF LONGWOOD, MASS. 
[Refrinted from The Boston Mkdicai. and Surgical, Journal, April 3. 1879.] 
CAMBRIDGE: 
printed at tf)c ftitoergi&e JDrc&& 
1879. 4 
3 
17 
c 
Note. In order to print these plates with greater distinctness, it has been necessary 
to arrange them in a different order from that in which they are referred to in the text. 
Heliotype Printing Co-., Boston. 15 
23 
1 
19 
13 
21 
Heliotype Printing CoBoston, 24 
2 
20 
18 
5 
.8 
Heliotype Printing Co., Boston. a 
22 
12 
27 
7 
18 
Heliotype Printing Co., Boston. 1<> 
28 
14 
20 
11 
Heliotype Printing CoBoston. EXPERIMENTS AND CLINICAL OBSERVATIONS 
ON THE 
IliEMATINIC PROPERTIES OF DIALYZED IRON.1 
BY ROBERT AMORT, M. D. 
Gowers’s hoemacytometer, the instrument I used in the following ob- 
servations, consists of a glass slide upon which are ruled squares one 
tenth of a millimetre in extent, and these are inclosed in a cell one 
fifth of a millimetre in depth. This slide can be used with any micro- 
scope, and with a lens of any desirable magnifying power; moreover, 
the magnified image of the slide may be projected by means of a pris- 
matic eye-piece upon a screen. The principle of Malassez’s “ compte 
globule ” does not admit of either of these convenient methods of delin- 
eation. In addition to the ruled slide there are two pipettes, one of 
which holds nine hundred and ninety-five cubic millimetres, and the 
other five cubic millimetres. The principle and method of using this 
instrument are then based upon the following facts : If a known meas- 
ure of blood be carefully drawn from an acupuncture, and intimately 
mixed with two hundred times its volume of a saline solution (specific 
gravity 1025) having a density sufficient to prevent the corpuscles from 
imbibing water, and so bursting their envelopes, these corpuscles, sep- 
arated from each other, will float in this mixture, and will finally settle 
down to the bottom of the containing vessel or dish. 
In this manner a small portion of the mixed or diluted blood is placed 
in the above-described cell, and the corpuscles are allowed to settle to 
the bottom, so as to be nearly on the same plane as the ruled lines, and 
are pretty uniformly distributed. The number of corpuscles in ten con- 
tiguous squares can then be readily counted, and this sum multiplied 
by the figure ten thousand will give the number of corpuscles in each 
cubic millimetre of the pure or originally drawn blood; because ten 
cubes, each one of which has the dimensions tVXtVXj millimetres 
(°r -glv of a cubic millimetre), will contain of a cubic millimetre, and 
since the original blood was diluted with two hundred volumes, the 
number of corpuscles actually counted in the ten squares must be mul- 
tiplied by two hundred times fifty, or ten thousand, in order to obtain 
1 Read before the Boston Society of Medical Sciences, January 21, 1879, and the Boston 
Society for Medical Observation, March 17, 1879. 2 
Hcematinic Properties of Dialyzed Iron. 
the number in each cubic millimetre of undiluted blood. If we wish 
still further to separate the corpuscles from each other, we may use four 
hundred volumes of the diluent to one of the blood, but in that case we 
must either count the corpuscles in twenty squares and multiply by 
ten thousand, or count those in ten squares and multiply by twenty 
thousand. 
The individual or personal error of vision which is associated with all 
optical instruments is perhaps somewhat difficult to reduce to mathe- 
matical accuracy, on account of the fact that constant observation fatigues 
the eyesight, and hence the results of a series of these observations are 
subject to an inconstant variation. In consequence of this apparent 
difficulty, I decided to project upon a photographic plate the image 
of the corpuscles on the ruled slide, then to print from the negatives, 
and count upon the print the number of these corpuscles, each one being 
obliterated as soon as counted. 
lo show the range of error of this method of obtaining the propor- 
tion of globular richness of the blood, I will call your attention to the 
prints Nos. 1, 2, 3, 4, which were all taken from blood of the same 
individual from four different punctures, separately diluted, and a sam- 
ple from each dilution separately photographed. 
No. 1 shows 4,656,000 corpuscles per cubic millimetre.1 
No. 2 “ 4,750,000 “ “ “ “ 
No. 3 “ 4,633,000 “ “ « “ 
No. 4 “ 4,673,000 “ “ “ « 
The greatest difference between these four results being only 117,000 
shows less than three per cent, of variation. Again, in three separate 
photographs, taken from one sample on the same slide, a field at either 
periphery, and at its centre, the variation was even less than that above 
mentioned. Thus, so far as this method is concerned, the error of vari- 
ation may be placed within three per cent. 
Hearing from certain members of our profession expressions which 
indicated a skeptical distrust in regard to the virtues of the solution of 
dialyzed iron, I was induced to undertake a series of observations upon 
the haematinic properties of this medicinal agent, and yet I must franklv 
acknowledge in advance that my own clinical experience with this form 
of iron had led me to attach to it a value in simple anaemia, and to 
place this remedy by the side of Quevenne’s iron. I have taken pains 
to inquire how much metallic iron is contained in the so-called dialyzed 
iron, and learn that a sample of the scaled dialyzed iron as prepared by 
one of our large wholesale druggists has been analyzed by Prof. Charles 
1 A reference to the prints will show not more than sixteen squares represented. On the 
original photographic prints presented at the meetings the count gave an average for twenty- 
five or thirty squares. To have reproduced this number of squares in each heliotype print 
would have required too much room. Hence a discrepancy between the numbers of corpus- 
cles mentioned in the text and what may be counted in the illustrations may be explained. Hoematinic Properties of Dialyzed Iron. 
3 
M. Cresson, of Philadelphia, and that he found it to contain fifty-two 
per cent, of metallic iron, or over seventeen per cent, more than is con- 
tained in the sesq.uichloride of iron ; compared with the sesquichloride 
it is equally soluble, keeps without change, does not corrode the teeth 
or fabrics with which it may be brought in contact, and contains a greater 
percentage of iron. The manufacturer from whom this sample was ob- 
tained claims that his solution of dialyzed iron has over twenty grains 
of ferric oxide in each ounce of solution, and in proof of this has shown 
me a certificate from Prof. F. A. Genth, of the University of Pennsyl- 
vania, that a sample of his manufacture, which was purchased by the 
latter from a retail druggist in Philadelphia, contained 21.69 grains of 
ferric oxide in each ounce of solution. 
The following table of assays of various iron preparations was made 
by M. Quevenne. One gramme (fifteen grs.) of the following prep- 
arations was soluble in two hundred grammes of gastric juice in the 
proportions named : — 
Of 1.0 iron by hydrogen containing 1.000 pure iron was dissolved 0.102 pure iron. 
Of 1.0 iron filings “ 1.000 “ “ “ “ 0.070 “ “ 
Of 1.0 protosulphate of 
iron “ 0.210 “ “ “ “ 0.056 “ “ 
Of 1.0 protocarbonate of 
iron “ 0.490 “ “ “ “ 0.050 “ “ 
Of 1.0 persulphate of iron “ 0.250 “ “ “ “ 0.046 “ “ 
Of 1.0 lactate of iron “ 0.190 “ “ “ “ 0.040 “ “ 
Of 1.0 protochloride of 
iron “ 0.430 “ “ “ “ 0.036 “ “ 
Of 1.0 tartrate of iron 
and potass “ 0.210 “ “ “ “ 0.022 “ “ 
Of 1.0 oxide of iron heated 
to red heat “ 0.700 “ “ “ “ 0.016 “ “ 
Forty parts of scaled or dialyzed iron from which water had been with- 
drawn without aid of heat was levigated very fine and placed in an artifi- 
cial gastric juice prepared as follows : five parts hydrochloric acid and 
acetic acid, fifteen parts of pure pepsin, with traces of chlorides of sodium, 
potassium, and ammonium ; also phosphates iron, lime, and magnesium 
in one thousand parts of water; and temperature was maintained at 
100° F. for five hours, then filtered through a “ tared ” filter, dried and 
weighed, showing a loss of iron amounting to fifteen parts. This shows 
that 3.00 grammes of scaled dialyzed iron (containing 1.5 grammes 
pure iron) are soluble in two hundred grammes of gastric juice. 
Having thus compared the properties of dialyzed iron with other well- 
known preparations, I will report five cases only, in which I observed 
its effect upon the globular richness of the blood. These were cases of 
simple anaemia uncomplicated by organic disease, and none of the indi- 
viduals were placed upon any regimen of diet or exercise, but were 
allowed to continue their usual habits of life. My object was to prevent 4 
Hcematinic Properties of Dialyzed Iron. 
the error of attributing any improvement in health or appearance to an 
improved diet, or to the more favorable effects of fresh air and exercise. 
I selected those cases in which I could personally observe the individ- 
uals from day to day, and could note any peculiarity of symptoms or 
unusual mode of life. The instrument which I imported from London 
was so inaccurate that I was forced to abandon it, as well as the reports 
of three cases in which it was used, and obtained from Prof. W. A. 
Rogers, of Harvard College Observatory, some very accurately ruled 
glass slides. I also procured some pipettes, and carefully estimated their 
capacity, so that I could place more reliance upon their measurements. 
I took especial care in the photographs herewith presented not to use 
various pipettes or ruled slides in the same individual; in the few cases 
in which (owing to accidental breakage) I was obliged to use more than 
one set of apparatus, the second set was carefully measured anew, so 
that my comparative results should not be invalidated. 
The circumstances under which the blood was taken from the patient 
were as nearly similar as possible, and I have a strong confidence that 
the results of my observations are not materially affected by the method 
of procedure, and that the error of variation may be safely placed within 
five per cent. I have not the time in this communication to prove the 
grounds for this statement, nor would it be worth your while to exam- 
ine the details of experiment which were used to determine my opinion. 
The pains that have been taken may perhaps be better appreciated by 
the careful examination of two hundred photographs, taken under vari- 
ous circumstances and conditions, among which may be mentioned the 
temperature of the surface after exposure to dry and wet heat and cold, 
the depth and method of puncture of the skin, blood taken while the 
patient was depressed by pain or temporary disturbances of other sorts, 
during the menstrual flow, and before rising from bed in the morning. 
The variation shown in these photographs is not to my knowledge de- 
pendent upon any of these disturbing causes. 
Case I. is that of a woman whose appearance was anasmic, and, as far 
as I could learn, had no organic disease. An earlier examination of 
her blood had shown about 3,900,000 corpuscles per c. mm. of blood, 
but this negative was unfortunately lost. The print No. 5 was taken 
on the third day after the commencement with the solution of dialyzed 
iron. I have estimated in this 4,189,000 corpuscles per c. mm., or 
about eighty-three per cent, of the normal state of health. At first 
she took her medicine quite regularly, and in ten days gained 171,000 
corpuscles, her blood having attained a percentage of eighty-six. The 
following six days she took the medicine very seldom, and lost 500,000 
corpuscles, and was in the same anasmic condition as before treatment, 
having a percentage of seventy-six. From this time onward she took 
the medicine more faithfully, but still omitting one dose about every Hcematinic Properties of Dialyzed Iron. 
5 
two or three clays, and in five days regained 440,000 corpuscles, having 
now a percentage of eighty-four. In another week she gained 60,000, 
and in four days more 160,000, having then a percentage of ninety-six. 
In the whole period, from December 11th to January 10th, this pa- 
tient showed an increase of 1,000,000 corpuscles, or an improvement of 
twenty per cent. During the whole time she was under observation 
she menstruated twice, but apparently with very slight effect upon the 
globular richness of the blood. None of these prints, Nos. 5, 6, 7, 8, 
9, 10, 11, were taken during menstruation. 
No. 5. December 5th, 4,189,000, or 83 per cent. 
No. 6. “ 13th, 4,360,000, or 87 per cent. 
No. 7. “ 19th, 3,850,000, or 77 per cent. (Has taken iron seldom.) 
No. 8. “ 24th, 4,240,000, or 84 per cent. 
No. 9. “ 31st, 4,630,000, or 92 per cent. 
No. 10. January 6th, 4,820,000, or 96 percent. 
No. 11. “ 10th, 4,860,000, or 97 per cent. 
Case II. is that of a girl about sixteen years of age, who consulted 
me for neuralgia, dizziness, especially during the act of stooping, lassi- 
tude, and general malaise ; she had the habit of awaking with a slight 
headache, which, towards the latter part of the day, forced her to leave 
her regular household duties. She had been out to walk two or three 
times a day, and, like all the other cases, had but slight loss of appe- 
tite, a good home, good food, warm clothing, without much require- 
ment for excessive mental or physical work. Her catamenia occurred 
too frequently, once in three weeks. Without advising any change in 
diet or occupation, I gave her in water, as in all the other cases, thirty 
drops of dialyzed iron solution .three times a day. I may as well say 
here that I obtained the solution of dialyzed iron always from the same 
manufacturer, furnished it directly to each patient, and that it had a 
uniform standard and purity, so far as I could determine. Its specific 
gravity was 1042, and in two instances the solid residue was about 
twenty-four grains to the ounce of solution. The record of Case II. 
is estimated from the prints : — 
No. 12. December 16th, 4,070,000, or 80 per cent. (Began treatment.) 
No. 13. “ 20th, 4,220,000, or 84 per cent. 
No. 14. “ 28th, 4,370,000, or 87 per cent. 
No. 15. January 6th, 4,620,000, or 92 per cent. (Discontinued treatment.) 
No. 16. “ 15th, 4,500,000, or 90 per cent. 
All her anaemic symptoms, above mentioned, gradually disappeared, 
and she omitted the treatment because she felt well, and up to the 
present time none of these symptoms have reappeared. 
Case III. is that of a young woman about twenty years of age, pale 
and thin in appearance, who has at previous times been much improved 
by a ferruginous tonic treatment. She complained of languor, dizzi- 
ness, muscular fatigue, cold, moist hands, painful menstruation, etc. 
Estimates from prints: — 6 
Hcematinie Properties of Dialyzed Iron. 
No. 17. December 17th, 3,700,000, or 74 per cent. (Began treatment.) 
No. 18. “ 31st, 4,700,000, or 94 per cent. 
No. 19. January 15th, 4,200,000, or 84 per cent. (Having a headache.) 
No. 20. “ 17th, 4,600,000, or 92 per cent. 
Apparently, on January 15th, her headache caused a diminution in 
globular richness, and for the sake of comparison I show print No. 21 
of my own blood, taken during a headache : ordinarily my blood shows 
over 5,000,000 corpuscles per c. mm.; the number estimated by this 
print is only 4,350,000. 
Case IV. is that of a clergyman, who had been suffering from neu- 
ralgia, headache, an uncomfortable feeling of pressure in the head, and 
general malaise ; these symptoms were aggravated after mental or phys- 
ical work. Though actively engaged in his professional duties, he led 
a healthy life, took daily exercise, eating well and at regular times. 
He had been taking a few weeks previously Carlsbad water, though he 
had no indigestion or constipation. His record is as follows: Omitting 
Carlsbad water, he began, December 3d, dialyzed iron solution, half a 
teaspoonful, diluted with water, three times a day, after meals, and taken 
regularly ; he omitted only one dose during five weeks’ treatment. The 
estimate from the prints is : — 
No. 22. 'December 3d, 3,800,000, or 76 per cent. 
No. 23. “ 13th, 4,500,000, or 90 per cent. 
No. 24. “ 28th, 5,191,000, or 103 per cent. 
No. 25. January 6th, 4,900,000, or 98 per cent. 
No. 26. “ 16th, 4,500,000, or 90 per cent. (The day after a headache, and iron 
omitted.) 
This was the only headache he had while taking the iron, and his 
professional work about Christmas time was excessive. 
Case Y. A married woman, who has had chlorosis and anaemia for 
several years, her appearance being white and bloodless, has been a pa- 
tient of Dr. Sabine’s for two years, and her chronic anaemia has been a 
marked feature ; apparently she has no organic disease other than this 
impoverishment of blood. Figure 27 indicates that the globular richness 
of blood is only 3,350,000, and at that time she began treatment by dia- 
lyzed iron. On one of the last days of February she had diarrhoea, 
with some catarrh of intestines, which was controlled in two or three 
days by ordinary doses of morphine and chalk mixture. On March 3d, 
after a fortnight’s use of the solution of dialyzed iron (Figure No. 28), 
her globular richness had improved to 3,560,000 in spite of the diarrhoea. 
On March L9th another examination showed 3,600,000 corpuscles per 
c. mm. 
There certainly is good reason for saying that four or five cases are 
insufficient to establish definitely that the solution of dialyzed iron cures 
anaemia. However, an analysis of the history of these cases shows that 
all the individuals were in comfortable circumstances, as far as food, 
clothing, and homes were concerned ; that none of them had any organic Hcematinio Properties of Dialyzed Iron. 
7 
disease (if we except simple impoverishment of the blood) ; that they 
were aware of being out of their usual health ; that they were unable to 
accomplish their regular work ; that none had impaired appetite, nor 
feeble digestion ; that three of the four suffered from neuralgia or head- 
ache ; finally, all had diminished corpuscular richness of blood, vary - 
ing from 3,350,000 to 4,000,000, and that under the continued use of 
ninety drops of solution of dialyzed iron per diem this condition of im- 
poverished blood was replaced by an increase in the number of cor- 
puscles, from 3,600,000 to 4,900,000, and the symptoms of ill health 
simultaneously disappeared with this improvement. 
Dialyzed iron may increase the globular richness of blood, but it may 
have no effect in bringing about the chemical combination of oxy-haem- 
oglobin. If the supposition be true that there is a state of ill health in 
which the corpuscles may be numerically normal, but may simultane- 
ously be deficient in coloring matter, we may conceive of an anaemic or 
chlorotic patient who may require some therapeutical means for improv- 
ing this deficiency other than simple iron. I may have been extremely 
fortunate in selecting just those cases in which a simple form of iron 
was indicated, and it may not be impossible that another form would 
have benefited my cases as much as the dialyzed iron; yet the latter is 
preferable to the more astringent iron salts, because it does not impair 
the digestion, nor produce constipation. 
I cannot close my communication without an expression of thanks 
to Dr. Sabine and my laboratory assistant, Mr. J. G. Hubbard, whose 
material assistance lent much to the value of the record of these experi- 
ments. 
Now, one final word about the various solutions of dialyzed iron. 
Many of these solutions are valueless, some are very dilute, and a few 
are of pretty uniform standard, and contain only the products of dialysis 
from a salt of iron and distilled water. If physicians use a worthless 
preparation, they need not expect an improvement in the anaemia; if 
they tise a dilute solution, they must prescribe a larger amount of the 
solution.1 In the preparation I used for these experiments the solution 
had a specific gravity of 1042, and had no free acid. 
1 See Bost. Med. and Surg. Journal, September 21, 1878.