the great death rate, which culminates in April; but this double decrease will soon change to less than average decrease, and then, as soon as there is an increase in sickness to more than average increase, culminating at the time of least deaths—which, in Michigan, seems to be in October, but in London is unmistakably in September then, as the deaths increase, the sickness under observation (old cases and new cases) should not increase as fast as it otherwise would. This may serve to explain why the curve for sickness drops from its maximum in April to its minimum in August, in four months, while it occupies double that number of months in going from its minimum to its maximum, which it does not reach until April, but which it might reach in March if the great number of deaths then did not keep it down. WHAT IS PROVED? It is useful to "take account of stock," as the merchants do at the close of the year, and ascertain just what is the state of our knowledge. In science I understand that a proposed mode of causation is considered proved when (1) it is demonstrated that the cause assigned is a vera causa—a true cause, capable of causing the phenomena ascribed to it; (2) that the cause assigned is present and acting, and (3) that no other known cause capable of causing the phenomena is present and acting. That chloride of sodium in strong solution is an irritant and a poison is a well-known fact, and the mode of death of animals poisoned therewith has been made the subject of experimental study.* That whenever a fluid containing a non-volatile salt is evaporated there is left a residue of salt, is now a part of our most common knowledge, that is implied by the term "non-volatile." That the blood serum and the fluids of the human body contain nonvolatile salts has been demonstrated by many analyses. That the quantity of vapor of water which air can contain is dependent upon the temperature is one of the most wellknown facts in meteorology, and that raising the temperature of air increases its capacity for moisture is equally well known. When cold air, which can contain only a small quantity of vapor of water, enters the air passages, and before it is exhaled is warmed, so that it can contain a large quantity, and is constantly in contact with moist membranes, from which it can take vapor of water-that it should take the moisture, and leave a residue of non-volatile salts, is in accordance with all our knowledge and experience on this subject. Finally, in pneumonia the increased non-volatile residue has actually been found, by analyses, in the sputa and in the solidified lung.† It seems to be demonstrated that the alleged evaporation of fluids containing nonvolatile salts takes place, and that the salt is "present and acting" in the air passages. (It does not change this fact if we admit that, normally, the residue left by evaporation is constantly reabsorbed or removed through the lymph channels; because it is possible that, when formed faster than normal, the lymph channels are not capable of removing the residue, or, if removed through them, the irritation may cause œdema sufficient to close those channels.‡) Considering that the temperature of the air exhaled from the air passages is always nearly the same, it would appear that the residue of non-volatile salts in the air passages should be quantitatively related to the temperature of the air inhaled—that is, to By B. J. Stokvis. By Lionel Smith Beale. Transactions Royal Medical and Chirurgical Society, London, England, 1852, Vol xxxv, pages 325-375. Interesting in this connection are the researches of Dr. H. F. Formad, of Philadelphia, which, he claims, prove that few and narrow lymph spaces in the connective tissue constitute a structural predisposition to tuberculosis.―Journal of American Medical Association, Vol II, page 148. In Archives für Experimentelle Physiologie und Pathologie Dr. Stokvis has shown that animals fatally poisoned by sodium chloride invariably have oedema of the lungs. the temperature of the atmosphere. That the sickness from several diseases of the air passages is quantitatively related to the temperature of the atmosphere is demonstrated by the statistics which I have presented. I know of no cause, other than the one I assign, capable of causing the irritation of the air passages, so as to control the rise and fall of coryza, influenza, tonsillitis, croup, bronchitis, pneumonia, and apparently also pulmonary consumption. For myself, therefore, it is proved (1) that the cause assigned is a true cause; (2) that it is present and acting when these diseases are caused, and that it is not only qualitatively but quantitatively related to these diseases; (3) that no other known cause is present and acting, even if we refer to each disease singly, and much less is there present any other known or alleged cause capable of inducing all of these diseases. It has been objected that it has not been proved that evaporation from the air passages extends to "parts so remote from the outer air as the pulmonary alveoli."* While it may be difficult to demonstrate this experimentally, I submit that it is susceptible of proof, as follows: Pneumonia rises and falls in relation to the atmospheric temperature in ways similar to those of the diseases of the upper air passages; if the diseases of the upper air passages are due to a non-volatile residue left by excessive evaporation, there is no other known cause "present and acting" to account for the pneumonia which is coincident therewith. Furthermore, the increased non-volatile residue having actually been found, by analyses, in the sputa and in the solidified lung of the person dead from pneumonia,† this alone amounts to nearly complete proof that the evaporation occurred, because (1) the evaporation of the fluid containing it is known to leave such a residue, and (2) there is no other known cause of the abnormal accumulation there of such a residue as the chloride of sodium. Finally, to assume that evaporation of moisture does not extend to the pulmonary alveoli is to assume that the air inhaled reaches its highest temperature before it reaches the alveoli, and that it is fully saturated with moisture at that highest temperature before it reaches the alveoli, neither of which assumptions is probably true, because the air comes most nearly in contact with the warm and fluid blood in the alveoli themselves. There may be no necessity for any further attempt to account for the fibrinous or albuminous exudations which occur in some or all of these diseases; the irritation of an abnormal proportion of non-volatile salts may be sufficient; but the fact that albumen will pass to a four per centum solution of sodium chloride may well be kept in mind in connection with this subject. I consider it proved, then, that the rise and fall of the diseases of the air passages are controlled by the atmospheric temperature, and that this is accomplished mainly through the quantity of vapor of water abstracted from the air passages. At the same time the mere evaporation of the water is probably harmless except it leaves an abnormal residue of non-volatile salts, which probably it may most readily do in persons whose blood is abnormally saturated with such non-volatile salts as usually pass out by way of the kidneys, and of which sodium chloride may be named as an example. THE READING OF THE DIAGRAMS. For the convenience of those who use the following diagrams, it may be stated that they are to be read with reference to the figures in the right- and left-hand margins, the numbers indicating the temperature being on the right and those representing the sickness or deaths, as the case may be, on the left. Thus, in Diagram No. 1 it will be seen that in the month of January, the average atmospheric temperature for ten years New York Medical Journal, Aug. 13, 1887, page 186. By Lionel Smith Beale. Transactions Royal Med. and Chirurg. Soc., London, England, 1852, Vol. xxxv, pages 325-375. was 20.56°, and in the same month the average percentage of reports which stated the presence of influenza was 55. In February the average atmospheric temperature was 23.62°, the percentage of reports stating the presence of influenza was 61. In August, when the curves for atmospheric temperature and sickness both reached their lowest point (the curve for temperature being reversed), the percentage of reports stating presence of influenza was 21, while the average atmospheric temperature was 68.14°. As an illustration, the table giving by months the average atmospheric temperature and the influenza (from which Diagram No. 1 is constructed) is given herewith: TABLE. STATING, BY MONTHS, FOR THE TEN YEARS, 1877-86, THE AVERAGE PERCENTAGE Percentage of weekly reports stating presence of Influenza....... Av. atmos. temp., degrees Fahr 55. 61. 59. 52. 38. 28. 20. 21. 29. 33. 41. 48. 20.56 23.62 29.80 44.33 56.08 65.10 70.52 68.14 61.67 50.83 36.04 26.60 For an exact reading of the figures, the tables which accompany the paper should be studied; but the relations of the temperature in one month to the sickness in that month or in a succeeding month can best be seen from the diagrams. In these diagrams, in which the unit of time is one month, and the curve representing sickness is made from reports of all cases under observation, old cases as well as new cases, the sickness curve should coincide with a curve representing a controlling cause of that sickness if the duration of the disease is less than one-half month, and the disease has no period of incubation, otherwise the curves may be separated by an interval corresponding, as nearly as the long unit of time will permit, to the average duration of the incubation and the sickness. |