Page images


E. H. B.

Entered according to Act of Congress, A.D. 1888, by A. N. BELL, in the office of the Librarian

of Congress, at Washington.

[ocr errors]



JULY, 1888.





By J. A. TANNER, M.D., Boston, Mass.

It is not my intention to discuss minutely the various methods for examining drinking water, or to attempt to point out any general or special procedure for examining and keeping pure the water supplies of cities, as the length of an article embracing such questions would be too great for this occasion. In view of the great hygienic importance of a supply of good drinking water, and the implicit confidence placed by many in the chemical examinations for the organic impurities in water, it will not be out of place to give a short criticism upon the value of these examinations in order that as physicians, not as chemists, we may form an idea of the ground we are walking upon when relying upon the reports based on such examinations.

Considering the multiplication of the sources producing waste products of a suspicious character, and the large increase of natural pollutions which tend to render impure the water supply of large communities, the question arises whether we are in command of a method for the examination of water that will give even an approximate idea as to its wholesomeness or unwholesomeness. Accepting as established the germ theory of the cause of disease—and the points brought forward in favor of it are strong--and in that connection viewing the recognized chemical methods used in examining water, it is

proposed to demonstrate that we are relying upon data received from unreliable processes.

Even those who do not accept the germ theory, when they contrast the gross amount of organic matter recognized by the chemical methods in comparison with the infinitesimal amount of infectious matter that will produce disease must question the value of such examinations.

Probably the most delicate test in water analysis is that used to detect the presence of the nitrite salts, and this, it is claimed, will recognize the presence of one grain of nitrites in one thousand million grains of water ; but this is a rough test compared to that necessary to recognize the exceedingly minute amount of septic matter that wili convey poison to a surgeon when he cuts his hand with a septic scalpel, or the syphilitic poison conveyed by the end of a cigar when the cigar has been rolled by a maker so diseased, who moistens his fingers with saliva when twisting the end. Again, the various infectious diseases carried by means of the air will pollute the water when this air is absorbed by it ; and as chemical examinations do not recognise this infectious matter in the air, it is not to be expected that they will detect it in water. It must be admitted that chemical processes will point out the presence of gross pollution and thereby give warning that the water may produce disease, but they do not give a definite idea as to the character of the organic matter entering into the pollution, nor do they tell us whether the matter is of that nature which will certainly produce disease under favorable circumstances; they do not recognize the difference between organic substances of a harmless nature and those of a highly dangerous character, though present in considerable quantities; and the processes will even pass by dangerous matter if present in small quantities. The quality of the pollution is the essential knowledge sought, not the quantity. Taking the processes generally recognized as giving reliable informationthe Frankland or Combustion process, the Wanklyn or Albuminoid process, and the Tidy or Permanganate process—and examining them in reference to the data furnished as to the potability of a water, it will be seen that it is entirely too meagre for definite work when applied to classifying, as to character, the organic matter that may be present.

To impress the idea to be conveyed, an illustration will be given. Suppose to a gallon of distilled water is added 5% of milk or any beef extract, and then this harmless solution is submitted to chemical examination ; the processes will condemn it as impure. Allow such a mixture to stand until the organic matter is decomposing, so as to render the water dangerous, and then examine it again; it will be condemned as in the first place. The data obtained from the two examinations will not be sufficiently different to point out any special difference between the two specimens, and in all probability the last examination will not condemn the specimen in terms as strong as the first did. The combustion process tells us that there is a certain amount of carbon and nitrogen present, and feebly attempts to classify the organic matter according to the ratio in which the carbon and nitrogen are found. The ammonia process, according to the presence of free ammonia and albuminoid ammonia, endeavors to name the character of the water pollution. The oxygen process relies simply upon the rapidity with which oxygen is taken up by the organic matter, in order to determine its character. If we will now invade the laboratory of the bacteriologist and manage to capture a few millions of some disease-producing bacteria, and add them to a gallon of distilled water, we will have, according to the germ theory, a highly dangerous mixture, but the chemical examination will pronounce this good, simply because there will not be sufficient organic matter present for the process to grasp and take notice of. Add a little fresh and innocent milk to this same water, and the processes will then condemn it, not on account of the bacteria present, but because of the organic matter contained in the milk. Passing by a discussion of the sources of inaccuracy due to the reagents, and the manipulations necessary in these processes, when we come to look at the arbitrary standard of classification, then the chances for error must be noted ; for we find the lines are drawn so closely that even, admitting the methods to be accurate in entity, from a very slight error in manipulation a water that is pure may be classed as impure, and vice versa. The permanganate process classifies a water according to the amount of oxygen that is required in oxidizing the organic matter present, say in one hundred thousand parts of water, as follows : A water re




[ocr errors]
[ocr errors]
[ocr errors]
[ocr errors]
[ocr errors]
[ocr errors]
« PreviousContinue »