120 Hemoglobin Determination 
tempted to overcome the objection by preparing a carbon mon- 
oxide hemoglobin standard (1 per cent solution of a blood hay- 
ing an oxygen capacity of 18.5 per cent) which he considered 
permanent when kept sealed in a small test-tube in an atmos- 
phere of carbon monoxide. 0.020 ec. of blood, placed in a simi- 
lar sized and shaped graduated (on scale of 100) test-tube with 
a small amount of water saturated with carbon monoxide by 
means of ordinary illuminating gas, are diluted with water, 
drop by drop, until the unknown and standard tubes match in 
color, whence the percentage of hemoglobin in the unknown may 
be read off on the graduated scale. In principle, Haldane’s 
method is sound; but certain practical difficulties arise. The 
standard is not so permanent as was at first thought, and, when 
water is used, it has been shown by Krogh (5) that the full color 
of the solution is slow in reaching its maximum. The further 
criticism is that the method is time-consuming and cumbersome— 
adding water, drop by drop, and shaking after each addition. | 
Also the color comparison in the two tubes is not sharp. ‘This 
method has one distinct advantage in that it may be used for the 
determination of hemoglobin in any species of animal. 
Sahli employed Haldane’s apparatus and technique; except 
that he used, as a standard, blood to which dilute hydrochloric 
acid had been added. There are three serious objections to 
Sahli’s method: first, the standard is not permanent; second, 
there is considerable delay in the development of the maximum 
er permanent color, amounting, according to Meyer and Butter- 
field (6), in some instances to 20 per cent; and third, it cannot 
be used for the blood of different species. 
The spectrophotometer has undoubted accuracy in the hands 
of skilled operators; but the expense and unavailability together 
with the difficult technique involved, make it impracticable for 
general use. 
The great variety of methods and apparatus which have been 
proposed offer eloquent testimony to the unsatisfactory means 
for the determination of hemoglobin. There is great need for a 
rapid, accurate, and universally standard method for the esti- 
mation of hemoglobin in experimental work and the study of blood | 
diseases in the clinics. 
The method which we have found to fulfil the above conditions 
is as follows: 
