National vitality, its wastes and conservation

Front matter

Gass T f ftVfiS Book

61st Congress \ SENATE I Document

2d Session ] \ No. 419

Section 1. — In different places

By those who have never considered the problem, death and disease are accepted as a matter of course. In individual cases it is recognized that a death or an illness might have been prevented, but the idea that the death rate could be changed in any appreciable degree, or controlled, is quite foreign to the mind of the average man. Charles Babbage wrote: "There are few things less subject to fluctuation than the average duration of life of a multitude of individuals.

If this statement were correct, we should find the average duration of life & and the death rate substantially the same in different places and at different times. The facts do not conform to this view. Modern life tables show that the average length of life in the leading countries of the world varies remarkably, as the following figures will illustrate/

Modern duration of life.,

Country.

Females.

England and Wales, 1891-1900

United States (Massachusetts, 1893-1897) a

• Samuel W. Abbott, M. D. : " Vital Statistics of Massachusetts for 1897," Thirtieth Annual Report of the State Board of Health of Massachusetts, 1898.

* Statistiques Generale de la France : Statistiques Internationales, etc., 1907, p. 566. See also J. A. Baines, " The Peradventures of an Indian Life Table," Journal Royal Statistical Society, 1908, Vol. LXXI, part 2, p. 310, where the average duration of life for males is given as 23.6.

When we consider that the average duration of life in India is scarcely more than one-half that of France and less than one-half

° Charles Babbage: A Comparative View of the Various Institutions for the Assurance of Lives, p. 15. London, 1826.

6 By average duration of life or the life span is meant the average length of life among a large number of persons born taken at random. It is not the same as the average age at death in a community during a year, since that community may contain an abnormally large proportion of infants, or of any other " age group."

c Some of the figures, especially those for India, rest on imperfect data, but they are believed to be sufficiently accurate for general comparison, that of Sweden, we must conclude that the length of human life is dependent on definite conditions and can be increased or diminished by a modification of those conditions.

Section 2. — At different times

Striking corroboration of this conclusion is found as soon as we compare the average duration of life at different periods of time. The earliest attempt to discover a law of human mortality appears to be that of Ulpian, a Roman pretorian prefect, about 220 A. D. The meaning of his table is somewhat doubtful, but it is assumed to refer to " expectation of life," which for a£es up to 20 is given as thirty years.*

This estimate is so crude and vague as to be of little value for comparative purposes. Professor Finkelnburg ? of Bonn, estimates that — -

The average length of human life in the sixteenth century was only between eighteen and twenty years, and that at the close of the eighteenth century it was a little over thirty years, while to-day it is between thirty-eight and forty years. 6

In Geneva the records go back over three centuries showing the following life span : c

16th century 21.2

rrth century 25.7

18th century 33.6

1801-1883 1 39.7

Here we see an increase in the span of life of 100 per cent in three or four centuries. The last few decades, moreover, tell a striking story of increase. It is one of the boasts of the nineteenth century that the splendid medical and scientific advances of that period have aided in a distinct lengthening of life.

In 1693 the British Government borrowed money by selling annuities, and in 1790, a century later, it did the same thing. While the first venture proved satisfactory, the second caused a great loss to the Government, owing to the improvement in longevity which had taken place, and which was estimated, for the annuitant class, at 20 years. d

If we compare Ogle's English life tables for 1871-1881 with those of Farr for 1838-1854, we find an increase in life span of 1.4 years for males and 2.8 for females.**

a See Irving Fisher : Mortality Statistics of the United States Census, Publications of the American Economic Association, Monograph on Federal Census, 1899, p. 157. Taken from Assurance Magazine, VI, p. 314 ; note.

& George M. Kober, M. D. : " Conservation of Life and Health by Improved Water Supply," publications of conference on the conservation of natural resources, 1908, p. 23. Washington, D. C., privately published. From Finkelnburg's " Organisation der offentl. Gesundheitsphlage in der Kulturstaaten " in Handbuch der Hygiene, 1893.

c From Mallet in "Annales d' Hygiene," XVII, 169 ; quoted by Dr. Edward Jarvis, " Political Economy of Health," in Fifth Annual Report, Mass. Board of Health, 1874.

d See Dr. Southward Smith, Transactions British Social Science Association, 1857, p. 498, quoted by Dr. Edward Jarvis, ibid.

e R. Mayo-Smith; Statistics and Sociology, p. 178. New York (Macmillan), 1895.

A still greater improvement has been effected since Ogle's figures of 1871-1881 :

Lifetime in England and Wales :

1838-1854 39. 9

1891-1900. 44.1

1838-1854 41. 8

1891-1900 47. 8

Similar improvements are observable in other countries.

Lifetime in France:

1817-1831 38. 3

1898-1903 45. 7

1817-1831 40. 8

1898-1903 49. 1

Lifetime in Prussia :

1867-1877 35. 3

1891-1900 41. 1

1867-1877 37.9

1891-1900 44. 6

Lifetime in Denmark :

1835-1844 42. 6

1895-1900 50. 2

1835-1844 44. 7

1895-1900 53. 2

Lifetime in Sweden :

1816-1840 39. 5

1891-1900 50. 9

1816-1840 . 43. 5

1891-1900 53. 6

It is difficult to obtain American life tables that go far enough back into history to display increases in the life span similar to those just presented; yet comparisons of Abbott's Massachusetts life tables for 1893-1897 with Elliott's Massachusetts tables for 1855 and Wigglesworth's Massachusetts and New Hampshire life tables of a century ago give us a progressive increase from 35 in 1789 a to 40 in 1855 b and 45 in 1893-1897.° Unfortunately no tables exist for the United States as a whole from which similar comparisons might be made. Good and reliable vital statistics are among our most crying needs. Meech's life tables, based on the census figures of 1830, 1840, 1850, and 1860, showed a life span for the whole country of 42. d

° E. Wigglesworth : " Table showing the probability of the duration, the decrement and the expectation of life in the States of Massachusetts and New Hampshire, formed from 62 bills of mortality on the files of the American Academy of Arts and Sciences in the year 1789." Memoir of the American Academy of Arts and Sciences, Vol. II, p. 133.

6 Proceedings American Association for the Advancement of Science, 1857, pp. 61 and 69. .

c Abbott, loc. cit.

<*Levi W. Meech: System and Tables of Life Insurance, ed. of 1886, pp. 255-259. The figures for life span at different periods given in this paragraph have been secured by averaging the figures for males and females.

The census of 1880 gave some 70 sets of life tables for different registration States and cities. The expectation of life for white males was given for Massachusetts as 44, New Jersey 46, District of Columbia 41, and New York City 33, but in constructing the tables " the census was too prodigal as to quantity and somewhat careless as to quality. It is difficult to separate the wheat from the chaff. The table should have been accompanied by a running criticism. The general defect was that no attempt was made to correct the deficiencies in the returns for infants." a The census for 1890 gives only a few life tables, and that for 1900 none.

In striking contrast to these recent increases of the life span in progressive countries is the table for backward India, which showed no advance in twenty years. 6

Lifetime in India : Males —

1881 23.7

1901 23.6

Chapter II. — The mortality rate. Section 1. — Relation of longevity and mortality

The average duration of life and the death rate are two complementary magnitudes. An increase in the life span means a decrease in the death rate, and vice versa ; in fact, in a " stationary " population (a population in which the annual number of deaths equals the annual number of births, and without emigration or immigration), it will be true that the average duration of life .and the death rate are mathematically the " reciprocals " of each other. d Thus, if the death rate is 20 deaths per annum for each 1,000 of population (i. e., twenty one-thousandths per annum), the average duration of life would be J^ - = 50 years.

If this reciprocal relation between duration of life and mortality held true in every population, it would be easy to translate death rates into average duration of life, and conversely ; but unfortunately such a simple calculation is impracticable under conditions existing in America, and even in most countries of Europe. With the exception of France, few countries have even approximately an equality between deaths and births and an absence of emigration and immigration. In America, where the deaths are exceeded by the births, and where there is a large immigration of young men and women in the prime of life, the death rate is smaller than it would be if our population were " stationary." The annual death rate in the United

a Fisher : Mortality Statistics of the United States Census, American Economic Association Monograph on the Federal Census, 1899, p. 160.

6 Baines : " The Peradventures of an Indian Life Table," Journal of the Royal Statistical Society, 1908, Vol. LXXL, pt 2, p. 310.

C A death rate is the ratio of the number of deaths during a year to the population, taken at some point in the year, usually the middle.

d For a short explanation of this reciprocal relation, which is more fully explained in works on actuarial science, see Fisher : Mortality Statistics of the United States Census, pp. 149-150. It is interesting to see the graphic interpretation of this reciprocal relation by means of the diagram in Chapter XI of this Report.

642 KEPORT OF NATIONAL CONSERVATION COMMISSION.

States is probably about 18 per thousand of population. The reciprocal of this would be 1,000-1-18, or 55 years, which is altogether too high an estimate for the average length or life in the United States.

It is possible to " correct " the death rate for " age distribution " so that its reciprocal will be the true average duration of life, but the calculation is a difficult and tedious one. We are forced, therefore, to get along in most cases with the " crude death rate," or the quotient of the number of deaths in a year divided by the population. This figure is much easier to obtain than a corrected death rate or its reciprocal, the average duration of life.

Our data for death rates are far more voluminous than our data for the average duration of life. Although theoretically death rates are not an unerring indication of longevity, they furnish in practice very valuable information. In a general way death rates may be compared with each other, especially in the same community. For instance, a decrease in the death rate in New York City from one year to another is practically a certain indication of improvement in vital conditions.

Section 2. — Mortality in various regions.

Forty years ago the variations in the death rates of the different sections of Europe were given by Quetelet b as follows :

Death rate per 1,000 population.

Northern Europe 24. 3

Central Europe 24. 5

Southern Europe * 29. 7

To-day the death rates of various countries compare with each other as in the following table :

Modern death rates per 1,000 of population.

England and Wales (1908) 15.4

United States (registration area) (1907) 16.5

As we found in the study of duration of life, so we find here wide variations from country to country. Italy presents a death rate larger by nearly one-sixth than that of the United States, while famine-tortured and plague-ridden India's mortality rate is twice that of France and three times the rates of Denmark and of Sweden.

W. F. Willcox : " Death Rate of the United States in 1900," publications of the American Statistical Association, Vol. X, p. 155.

6 Quetelet, Physique Sociale, Vol. I, p. 281. These figures refer to the experience of different periods between 1801 and HS31.

c With the exception of the figures for India, this table is taken from figures furnished by the Bureau of the Census.

d Baines, loc. cit, p. 310.

Even the fairly homogeneous population of our registration States in America shows variations in death rates.

Death rate per 1,000 population in 1900.

Michigan 13. 9

Vermont 17.

Massachusetts 17. 7

New York 17. 9

The death rate in Michigan, at the one extreme, is thus but three-fourths that of New York at the other extreme. This difference may probably be accounted for in part by the difference in the age constitution, as the population for Michigan contains a larger proportion of population in young and vigorous life than does New York.

Section 3. — Urban and rural mortality. Comparison of urban and rural death rates also gives us variations. 6

Death rates per 1,000 population m 1900.

Massachusetts :

Urban 17.9

Rural 17. 1

Michigan :

Urban 15.3

• Rural 13. 3

New Jersey :

Urban 18. 8

Rural 15. 5

Interesting comparisons may be made of the death rates of American cities varying in size and location. The death rate per 1,000 of population in 1906 was given c as 14,2 (probably incorrect) in Chicago, in Boston 18.9, in New York 18.6, and in Philadelphia 19.3. Cleveland, Ohio, was credited with a death rate of but 16, while Cincinnati, in the same State, had 20.8. The causes of such differences are not always attributable to variations in size. New Haven, for instance, a larger Connecticut city than either Hartford or New London, had a lower death rate in 1900 by 2.2 and 2.5, respectively, per 1,000 population. The differences are accounted for partly by differences in age constitution, partly — it is unfortunately true — by differences in the accuracy of the collected statistics, partly by differences in size and location, and partly by differences in the vigilance of the public and private health authorities.

European cities show even greater variations in mortality than these just given for the United States.

Twelfth Census of the United States, Vol. Ill, p. LVIII.

6 United States Census Bulletin, No. 83, 1901.

c Seventh Annual Census Report on Mortality Statistics, 1906.

Death rates of European cities per 1,000 population, 1897.°

Locality.

High.

Locality.

Low.

Dublin

39.9 36.9 31.3 31

Frankfort on the Main

15 6

Moscow

The Hague

16 2

Belfast

Berlin

St. Petersburg

Amstfirrla;m

17.8

° One must doubt, however, the accuracy of some of these figures, especially the Russian. The rates in Moscow and St. Petersburg in 1906 are reported as 25.8 and' 25.5, respectively, which do not comport with the rates for 1897.

Section 4. — Race and condition

The variations in death rates among different races are well known. The black race, for example, always suffers a higher mortality than the white. In Boston during the half century from 1725 to 1774 the death rate per 1,000 is given as ranging from 56 to 87 for the blacks and only from 30 to 41 for the whites. Thus the maximum white death rate was lower than the minimum black death rate. a

In 1906 the death rate per 1,000 in all registration cities having not less than 10 per cent colored inhabitants was 17.2 for whites and 28.1 for blacks.

These racial differences may be ascribed in part to different habits and conditions of life, but probably in part also to varying racial susceptibility to disease.

The relation of social status to the rate of mortality has been often discussed and offers a partial explanation of racial or national variation of death rate. That a well-to-do class, properly fed and clothed and with opportunity for leisure, will be less susceptible to disease and death than a poverty-stricken class, ill-fed and overworked, has been repeatedly shown by statistics. Newsholme has stated, 6 for example, that in Glasgow the death rate among tenants of large houses is much lower than among the tenants of smaller dwellings :

One and two room houses.

Three and

four room

houses.

Five rooms and over.

Death rate per 1,000 occupants in 1885

27.7

19.5

11.2

In Paris comparison has been made between two quarters known to be rich, on the one hand, and, on the other, a third quarter known to

be poor.

Death rate per 1,000 population.

Rich quarters:

Elysee 13.4

Opera 16. 2

Poor quarter :

Menilinontant 31. 3

Lemuel Shattuck, " The Vital Statistics of Boston," p. xiii. Reprinted in " Bills of Mortality, 1810-1849," city, of Boston, 1893.

6 Arthur Newsholme, "Vital Statistics," London (Swan Sonnenschein), 1S99, p. 163.

c E. Levasseur, " La Population Francaise," Paris, 1891, Vol. II, p. 403.

FISHEE.]

NATIONAL VITALITY.

In Eussia a similar comparison has been made between peasants who own no land, those who own less than 13J acres, those who own between 13J and 40J acres, and so on up the scale of proprietorship.

Peasant death rate per 1,000, Government of Voronezh, 1889-1891.

Class of household.

Having no land

Having less than 13.5 acres

Having 13.5 and less than 40.5 acres Having 40.5 and less than 67.5 acres Having 67.5 and less than 135 acres Having more than 185 acres

Occupational comparisons are often made; and while they must be handled with great care, especially because of differences in age, the following may be said to display roughly the variations in death rate among social classes.

Death rate of males per 1,000, according to occupations, for registration States,

1900*

Mercantile and trading 12. 1

Clerical and official 13. 5

Professional 15. 3

Laboring and servant 20.2

Special industries have high death rates from special diseases. Among dusty trades, for instance, tuberculosis is very common.

Finally, the experience of industrial life-insurance companies, which deal largely with the poorer classes of society, shows a higher death rate than that displayed by the experience tables of other insurance companies. 5

Insurance mortality per 1,000.

Ag«

25, 35,

55 70,

Ordinary

insurance,

English

experience.

Industrial insurance, Metropolitan Life.

We find also great variations in death rates dependent on. varying climatic or seasonal conditions, on the prevalence or absence of certain pests, on the fluctuating virulence of specific diseases, and on numerous natural differences. Other significant factors in mortality are historical events, such as wars, plagues, and epidemics. Hard times bring increased mortality, whether due to natural or politico-economic causes. There remain to be mentioned also deaths by accident in all its many forms.

I. M. Rubinow, " Poverty's Death Rate," publications of the American Sta- .tistical Association, December, 1905, p. 348.

6 " Twelfth Census of the United States/' Vol. Ill, p. cclxi.

c See Frederick L. Hoffman, " Mortality from Consumption in dusty trades," Bulletin No. 79 of the U. S. Bureau of Labor.

d Haley Fiske, " Industrial Insurance," The Charities Review, March 1898, p. 33.

646 EEPOET OF NATIONAL. CONSEBVATICN COMMISSION.

Section 5 — Mortality historically

Not only does the death rate vary greatly from place to place and from one social class to another, but it changes in a most marked fashion from period to period in history. The records of old cities show that a decided decrease in mortality has been steadily going on. In London, for example, the rate per 1,000 has fallen from 50 in 1660- 1679 to 15 in 1905, a decrease of 70 per cent. In the plague years, 1593, 1625, 1636, and 1665, the death rates per 1,000 were 240, 310, 130, and 430.° The " black death " in 1348-9 probably swept away half of the population in many localities throughout Europe. 6

Mortality in London.

Year.

Rate per 1,000.

1660-1679

1729-1780

15.1

• Kober, loc. cit., p. 25.

•These years include the period of pests.

Within a quarter century London has cut her death rate in half, while Vienna, if we may trust the figures, has within a century reduced her rate from 60 per 1,000 of population to 23. c Similarly, the mortality rate in Boston has been lowered from an estimated 34 per 1,000 d in 1700 to 19 to-day.

Mr. John K. Gore, actuary of the Prudential Insurance Company, shows e that the average death rates per 1,000 of population among typical American cities was, for the white population, as follows:

Years. Death rate per 1,000.

1804-1825 24. 6

1826-1850 25. 7

1851-1863 28. 3

1864-1875 25. 4

1876-1888 22.9

1889-1901 21.0

The record even of the last thirty years displays a fall in death rates that may inspire us with buoyant hope for the future. The mortality rate per 1,000 has fallen in Berlin from 33 in 1875 to 16 in 1904; in Munich, from 41 in 1871 to 18 in 1906; and in Washington, from 28 in 1875 to 19 in 1907/

Between 1890 and 1906 New York lowered her death rate per 1,000 from 25.4 to 18.6, and Boston from 23.4 to 18.9. The mortality rate in the whole registration area of the United States fell from 19.6 per 1,000 in 1890 to 16.1 in 1906, although the area in the last-named year included a larger proportion of urban population.

° See Farr, Vital Statistics, London, 1885, p. 131.

& See Abbot Gasquet, "The Black Death of 1348 and 1349," London (Bell), 1893.

"A. F. Weber, " The Growth of Cities," New York, 1899, pp. 355, 356.

d Shattuck, loc. cit., pp. xii-xiv. '

e " On the Improvement in Longevity in the United States during the Nineteenth Century."

We have also vital records for the city of Habana, running back over a century. These show that while the death rate in 1802 was given as 54.6 per 1,000, rising in cholera years even as high as 103.4 (1833) and in the last year of Weyler's concentration methods as high as 91, the rate during the eight years from 1899-1906 ranged from 20.4 to 33.6.°

These records also show the remarkable and sudden fall that may be brought about by a change in the living conditions of a community. During the three " concentration " years of 1896, 1897, and 1898 the mortality rate per 1,000 was 51.7, 78.7, and 91, respectively. In 1899, the first complete year of American occupation, the rate dropped to 33.6, and since then it has ranged between 20.4 and 24.4. There can be no question that the improvement was almost wholly due to the sanitary reforms introduced by Colonel Gorgas, and the other United States Army surgeons under Gen. Leonard Wood.

The record of American Army sanitarians in the Panama Canal Zone shows as striking results as in Cuba. The death rate in Panama during 1887, when the French canal companies held occupation, ran over 100 per 1,000. In 1906 the death rate was 49 per 1,000, while in 1907 it fell to less than 34. Colonel Gorgas attributes the decrease in the general death rate in great part to improved sanitation, though he adds that " increased wages, better food, and better clothing have no doubt played a considerable part in the general improvement of the health." b

Section 6. — Adult and infant mortality. '

Mortality varies greatly with age. The improvement in the city death rate of the past half century has been especially marked among the young.

It is true that in countries of the same degree of civilization the infant death rate is remarkably constant, but this is probably accounted for by the similarity in the methods of feeding of infants. Certainly where there is a difference in conditions there will be found a difference in mortality. Thus, the comparison between the mortality of infants fed on cow's milk and those fed on mother's milk shows that the former is five to ten times that of the latter.**

Although the infant mortality rate is probably falling, e the de 1 crease is not accompanied by a lowering of the mortality of later life. There is an increased mortality beyond the age of 50 years. In Massachusetts the death rates by age changed during thirty years as follows:

° " Report of the National Sanitary Department of the Republic of Cuba," Habana, 1906, p. 78.

& Letter from Col. W. C. Gorgas, chief sanitary officer, Canal Zone.

c See E. B. Phelps, "A Statistical Study of Infant Mortality.'* quarterly publication of the American Statistical Association, September, 1908. Mr. Phelps shows the utter unreliability of most statistics of infants under 1 year of age, especially in the United States.

d See Harald Westergaard, " Mortalitat und Morbilitat," Jena, 1901, p. 364.

e Edward E. Graham, M. D., " Journal of the American Medical Association,'*' September 26, 1908 ; also " British Medical Journal," February 1* 1908, p. 271.

Death rate m Massachusetts per 1,000 of population in each age period. 9

Age.

6-9 ;.,

9.6 5.1 9.1

12.6

11.7

70 168

6.2

10-14

3.2

15-19

5.3

20-29

7.1

30-39

9.7

40-49

50-59

60-69

70-79

80 and upward

« " Vital Statistics of Massachusetts," 1856-1895, p. 755.

Here, while the death rate for all age periods under 40 has materially decreased, the later periods of life have suffered progressive increases in mortality rate.

As Frederick L. Hoffman has expressed it :

There is, of course, no question whatever that the American death rate, using the term in a very comprehensive sense, has substantially declined within the last fifty years, but it is equally evident that this decline has been at the younger ages, and not during the period of life which, economically, is of the greatest value. There is no doubt that the mortality of adult ages is still decidedly excessive.

The same tendency, viewed from the standpoint of the expectation of life, 6 is disclosed in the study of two Massachusetts life tables, compiled nearly a century apart — one, Wigglesworth's life tables for Massachusetts and New Hampshire in 1789, though not very accurate ; the other, Abbott's Massachusetts life tables for 1893-1897.

Expectation of life in Massachusetts.

Age.

Age.

35.5

43.2 34.2 26

45.4

28.2

15.4 5.9

3.7

15.1

10 ;

6.1

3.4

.These figures indicate that the expectation of life at the earlier ages is much greater than a century ago, but that for the age of 60 and upward it has remained practically stationary.

°At first sight it would seem that this increased mortality in later ages could be explained away as due to the larger number of persons who are saved from earlier death and tend to produce a higher mortality at the older ages. I speak of this because very intelligent persons have drawn this conclusion. But it is obviously fallacious, since the figures do not indicate the number of deaths in different periods, but the death rates in different periods of life per 1,000 at each period. There are, to be sure, as a consequence of saving lives in the past, more old men now living than otherwise there would be, and there will be more deaths, but the figures show that these old deaths have increased faster than the number of old men. This fact raises the suspicion, therefore, that the lives which have been saved by the hygiene of a generation ago are weak lives.. Whether this is a tenable hypothesis or not will be discussed in Chapter X.

6 " Expectation of life " at any given age is the mean after-lifetime of persons who reach that age. Thus 100 persons have an expectation of life of 50 at the age of 10, if the total life to be lived by those 100 persons before death is five thousand years.

NATIONAL VITALITY.

English life tables ° for three decades ending 1900 display the same tendency.

English life tables — Expectation of life.

MALES.

Age.

1871-1880.

1881-1890.

1891-1900.

41.4

39.4 13.1

4.8

43.7

40.3

12.9

4.5

44.1

12.9

4.6

FEMALES.

44.6

41.7

14.2

5.2

47.2

42.4

14.1

47.8

43.4

14.1

5.1

These tables show that there is improvement at the younger ages for the period 1891-1900 over the period 1871-1880. For ages over 60 there has been a retrogression. It is observable, however, that between the periods 1881-1890 and 1891-1900 the figures for 60 years have remained stationary, and for 80 have slightly improved.

In other words, a baby to-day has in prospect a much longer average lifetime than did the baby of two generations ago; but a man or woman 60 years old has in prospect an average after lifetime no greater than formerly.

The proximate cause of this contrast would seem to lie in the fact that the mortality from many of the diseases of later life has been and is on the increase. The death rates from diabetes, heart disease, and Bright's disease have all doubled in thirty years. 6

Cancer is probably on the increase, and " to-day one in every 21 men who have reached the age of 35 and one in every 12 women who have reached 35 eventually die of that disease." d

In addition, there may be mentioned other diseases, arteriosclerosis, nephritis, apoplexy, paresis, disorders of the liver, and all manner of degeneration, all of them maladies of adult life, and all of them apparently tending to increase.