Alcohol and Life Insurance

I

APPROXIMATELY two thousand million gallons of alcoholic beverages are consumed each year in the United States. This whole vast aggregate is created from materials such as starch and sugar, which are necessary for the sustenance of mankind. Prodigious quantities of good food, soundly adapted to the human machine, are expensively treated and destroyed to make alcohol. Surely there is matter here for consideration in connection with the high cost of living.

But the alcohol question far transcends consideration of expense. In considering its importance to humanity, our inquiry naturally separates into three lines of investigation. First, there is its food value, because for ages alcohol has been used on the table. Rightly or wrongly, it has found its way into the good society of nourishment and hospitality. Second, its social value, its peculiar effect on brain and nervous system, entirely apart from its energyproviding or tissue-sparing qualities. Third, the demands it makes on the pocket and body, on mind and life.

These questions are of enormous social and economic importance. If real happiness, even though fleeting, if real energy, though only for the creative moment, can be secured for the price of a drink, let us all drink! Let us teach our wives and children to utilize these blessings of alcohol, if such there be, while also teaching its dangers. If the steady use of alcohol is beneficial to normal people, let us convert the total abstainer from his evil ways of neglect, and teach him the great benefit of indulgence. Let us warm him up and civilize him; let us show him what he is missing in life. But bear in mind that the question is put in a rigidly literal sense: Can real happiness and energy be secured for the money price of a drink ?

This is a matter to be settled by logic applied to evidence. Unfortunately, as Spencer has said, the judgments of most men, even on the more important affairs of life, are determined rather by feeling than by evidence. This is peculiarly true as to the alcohol question.

As Dr. Haven Emerson recently expressed it, it is difficult for the physician who has seen the ravages of alcohol among hospital and dispensary patients to be impersonal and impartial in discussing it. Its ill effects are so frightful, the misery it unquestionably causes so widespread, that the impulse is to banish it forthwith from the earth rather than to take any further chances with its effects on humanity.

On the other side, however, we have the testimony, not to be lightly treated, of those who have used it temperately and value it; whose ancestors have used it temperately and valued it; who regard it as one of the solaces of mankind in a world that is all too full of weariness and sorrow and fatigue; who look upon it in part as a food, in part as a useful sedative to over-active or strained faculties which are battling with an existence daily growing more complex: such people would agree with John Fiske, that alcohol ‘ diminishes the friction of living’ and ‘bridges over the pitfalls which the complicated exigencies of modern life are constantly digging for us.’

Regulation and moderation, rather than total abstinence, appeal to such thoughtful people as a reasonable compromise with alcohol. There are others — a very large percentage — who do not think at all, but just drink, and drink, and drink.

Those who are seeking to come to some conclusions as to the ultimate functions of alcohol and the desirability of its restriction to inanimate machines whose insides it cannot injure, would do well first to measure its total effects on humanity. Why go through the throes and agonies of debate and propaganda until we have settled the question as to whether or not the socalled moderate drinker needs to be disturbed in his indulgence, either by education or by legislation?

In a strictly scientific sense, of course, no drinking is moderate that causes any injury to the body, however slight, or that in any way impairs the efficiency of the mind or body. If alcohol in the smallest doses usually taken produces injurious effects and in any way lessens the efficiency of the body, then there is no such thing as moderate drinking. The question, Does moderate drinking shorten life? is a contradiction in terms. The logical form of the question is, Does alcohol in the smallest quantities customarily used as a beverage shorten life, impair the efficiency of the human body, or in any way adversely affect the mind, character, or career of the user?

What then are the sources of evidence?

Common observation of the obvious effects of fairly free indulgence in alcohol is not without its value. Every Sensible person recognizes that there is a danger-line in the use of alcohol long before visible intoxication is reached, and that life and character and action may be unfavorably affected by steady drinking. The most comprehensive and exact source of information, however, should be the death-rate among large masses of human beings who have used alcohol, as compared with other large masses, alike as to age and general condition, who have not used it. Fortunately, there is a large body of important evidence on the subject, which has lately been much increased in volume and in accuracy by the investigations of American life insurance companies.

In addition to the experience on large groups of lives as regards mortality, and the incidence of special forms of disease, we require the testimony of the laboratory as to the effect of alcohol on individuals — the measured effect of small doses, or at least doses approximating the smallest amount that would be used as a beverage. What action lesser amounts may have is a purely academic question, not confronting us in the workaday world. We also need the testimony of the clinic and of the sick-room as to the apparent effects of alcohol in therapeutic doses, its apparent influence on the course of disease. This latter evidence is less exact, yet not without its value when combined with evidence from other sources.

Now, the purely statistical evidence derived from life-insurance experience must be interpreted in the light of the evidence from varied sources on the effects of alcohol on the human organism. To accept the life-insurance figures at their face value would be unscientific and unsafe. Furthermore, the life-insurance statistics must also be considered in the light of evidence as to how they were accumulated, the various factors entering into the selection of insured lives that might influence the mortality in the classes considered.

The first important compilation of life-insurance experience showing the comparative mortality of abstainers and users, was that of the United Kingdom and General Provident Institution of Great Britain. This institution was founded at a time when the total abstainer was looked upon as a ‘queer duck,’ probably mentally unbalanced and certainly physically weak. In fact, this particular company was founded by a man who had been asked to pay an extra premium because he insisted on being a total abstainer. Later in the history of the company both classes were accepted, abstainers and users, but in no case those who were other than temperate in the use of liquor.

The records of this institution show that during the period from 1866 to 1910, the users of alcohol, who were about equal in number to the non-users, and selected as high-grade, temperate ‘risks,’ showed a mortality 37 per cent in excess of that exhibited by the abstainers. These figures are particularly significant because the users of alcohol showed a lower mortality than that of British insured lives generally. Thus the comparison was made with a high standard of excellence, and yet the abstainers excelled the standard to a marked degree.

It is of interest to know that, while in the course of the company’s whole experience the excess mortality among users was 37 per cent, the mortality among users between the ages of 35 and 40 was 83 per cent in excess, showing the influence of some extremely unfavorable factor at that critical period. Was it alcohol? Let us wait and see.

The best evidence of the reality of the saving among the abstainers is the action of the company in paying bonuses to the abstainers, and the action of other companies in anticipating such bonuses by charging lower premium rates to abstainers. The saving is just as genuine and positive as that affected by a manufacturer who is able to reduce the cost of manufacturing hats, or automobiles.

Other British and Scottish companies — the Sceptre Life, the Scottish Temperance Life, the Abstainers’ and General Life — show similar experiences; not over so long a period or for such a large group, to be sure, but the evidence is all in one direction, the excess mortality among users ranging from 37 per cent to 51 per cent. These figures, though widely quoted by temperance reformers, have not until recently been accorded their full value by actuarial authorities in this country. Various explanations other than the influence of alcohol were offered, and the fact that they were chiefly British figures deterred many from accepting them as reflecting probable conditions among insured risks in this country.

Ekholm, reporting the experience of the Swedish Life Insurance Company, 1897-1906, on about 35,000 lives, of whom somewhat more than half were abstainers, found a general difference of 6 per cent in favor of abstainers, even in that brief period. Beyond the age of 44 the difference was 26 per cent in favor of abstainers. Here, too, the trend of the mortality (the difference in favor of the abstainers increasing with increasing age of the policies) suggests that on an experience as extended as that of the British company the abstainers would show an advantage of approximately the same degree.

Ekholm, again quoted by Quensel in the Year-Book of the United States Brewers’ Association, claims that, to settle the question properly, comparison should not be made of abstainers with users; but that the mortality among those using alcohol in varying degrees should be compared with that among abstainers, so that zero might be compared with slight indulgence, moderate indulgence, free indulgence, and so forth. He states that such statistics are now being collected in Sweden, but that it will take a long time to complete them.

Fortunately, we have in this country comprehensive evidence exactly along these lines. In 1908, forty-three American life insurance companies — the older established, ‘old line’ companies, few of which had made any important effort to establish a total-abstaining class — undertook to investigate their mortality experience among various classes of ‘border-line’ risks, such as overweights, underweights, those with histories of various illnesses, and the like, and included in the investigation the mortality among the users of alcohol, classified according to the degree of their indulgence. No investigation was made of total abstainers. There was also computed the relative deathrate from such organic diseases as pneumonia, Bright’s disease, and cirrhosis of the liver among the various classes of drinkers, and among those engaged in the various branches of the liquor trade.

II

The investigation covered the period between the years 1885 and 1908, and its great value lies in the fact that the material was drawn from the records of 2,000,000 policyholders, and that all individuals were excluded from the study except those of sound average condition when insured. The final groups studied were, therefore, pure, or homogeneous, except for their varying use of alcohol or their varying exposure to alcohol as determined by their occupations. All extraneous influences, such as overweight, underweight, impaired family history, or personal history, were excluded. The results may be subdivided as follows.

First, those who were accepted as standard lives, but whose histories showed occasional alcoholic excess in the past. The mortality in this group was 50 per cent in excess of the mortality among insured lives in general, equivalent to a reduction of over four years in the average lifetime of the group.

Second, individuals who took two glasses of beer, or a glass of whiskey, or their alcoholic equivalent, each day. In this group the mortality was 18 per cent in excess of the average.

Third, men who indulged more freely than the preceding group, but who were considered acceptable as standard insurance ‘risks’. In this group the mortality was 86 per cent in excess of the average.

It should be borne in mind that these comparisons are made with the general class of insured individuals, both users and non-users of alcohol. Comparison with total abstainers alone would probably show much greater differences. It is noteworthy that in these drinking groups the death-rate from Bright’s disease, pneumonia, and suicide was above the normal, and that among the steady so-called moderate drinkers — those using more than two glasses of beer or one glass of whiskey daily — the death-rate from cirrhosis of the liver was five times the normal.

The story becomes monotonous in its uniformity: wherever we find alcohol we find higher mortality, and a mortality consistently increasing with an increasing use of alcohol. But we must follow on the story to the end.

The collection of these insurance statistics was made under the direction of a committee of actuaries and medical directors selected by the representatives of the important life insurance companies of this country because of their peculiar fitness for the task. This committee, of which Mr. Arthur Hunter was chairman, constituted an extremely conservative as well as highly specialized body of intelligence; indeed, these men were fairly saturated, not only with scientific, but with business conservatism, for their main purpose was to ascertain facts from which principles of business action could be derived. Alcohol was merely one feature among many that were investigated, and the conditions were such that no possible bias could influence the results. Nevertheless, these figures have been attacked by some and belittled by others, on the grounds that they were loosely assembled and that the varying practices of the contributing companies in accepting risks tended to make the results misleading and interfere with a well-balanced study of the conditions.

To test the general consistency of the figures, Mr. Hunter analyzed the experience of the company of which he is actuary, the New York Life, with the following results: —

Those drinking beyond 4 Anstie’s limit’ (one and one-half ounces of absolute alcohol daily¹) were classified as steady free users. The mortality in that class was double that among the general body of policyholders accepted at regular rates: that is, the extra risk on these lives was as great as in cases of heart disease, syphilis, and other diseased or unpaired states that loom much larger in the mind of the average drinker than does his indulgence. In fact, the insurance company treated these cases accordingly, and placed liens on the policies, or charged ‘rated up’ premiums, to provide against the expected high mortality. This was ‘cold business’; no bias, no sentiment, but a charge for the risk, just as would be the case in an extra fire or marine risk. The man who passes Anstie’s limit goes into a sub-standard class, and a poor sub-standard class at that. While individually he may escape, he belongs to a class that is fated to lose twice as many men in the same space of time as the general average.

Further figures from the same company show the following: —

Approximate Extra Mortality

Excessive use of alcohol a short time prior to date of application. 80 per cent

Excessive use of alcohol not recently, but within 5 years of date of application ... 45 per cent

Entire class with history of excess, including above and also those whose last excess occurred more than 5 years prior to date of application. 60 per cent²

This confirms the clinical judgment expressed recently by Dr. Bernard Sachs, the neurologist, at a meeting of the New York Academy of Medicine, that alcoholic excess always leaves some permanent injury on the life.

The figures of the New York Life Insurance Company show that among those who had taken a cure for alcoholism, but were temperate at time of acceptance, the extra mortality was 70 per cent. In the experience of the forty-three companies, among those who had taken a cure, but remained total abstainers up to the time of acceptance, the mortality was 35 per cent above the normal. Those who had been heavy drinkers, but who had reformed without taking a cure, showed an extra mortality of 32 per cent, doubtless because of stronger will-power and sturdier nervous constitution. These comparisons are made, as has been stated, with the general class of insured lives. In order to measure more closely the effect of socalled moderate drinking, Mr. Hunter has recently brought together a considerable group of experiences in which the relative mortality among abstainers and users in a number of American companies is shown.

The Manufacturers’ Life of Canada, the Security Mutual of New York, the Mutual Life of New York, have all published experiences which have now been brought up to recent date, and in addition the results in the Northwestern Mutual, the Phoenix Mutual, the Peoria Life, the Equitable Life of New York, and the New England Mutual are shown. Through Mr. Hunter’s kindness I am able to produce these figures.

The Mutual Life’s experience from 1875 to 1889 showed a mortality among abstainers 23 per cent less than among users. It was also found that the deathrate among beer-drinkers was almost the same as among wineand spiritdrinkers. A second investigation by Dr. W. E. Porter, medical director of the company, covering the years 1907 to 1912, showed the mortality among total abstainers to be 17 per cent less than among temperate, and 26 per cent less than among moderate, users. These are doubtless minimum figures, as they cover only a few years, and do not give time to reckon with the full adverse effects of alcohol, especially as the cases under investigation had only recently been accepted after careful medical examination.

In the New England Mutual Life, Dr. Dwight reports the mortality among abstainers to be 17 per cent less than among those rarely using, 29 per cent less than among temperate, and 53 per cent less than among moderate, users. These classes were fixed by the habits at time of acceptance for insurance. The results were almost parallel with cases of tobacco-users, except that the free users of alcohol showed a much higher mortality than free users of tobacco.

The Northwestern Mutual’s experience exhibited some curious conditions. Temperate beerand wine-drinkers showed a mortality only about 3 per cent in excess of abstainers, while whiskey-drinkers and heavy beer-drinkers showed a death-rate 25 per cent higher than the abstainers. This company had a clause reserving the right to cancel a policy for intemperance, and there was no attempt made to determine whether the abstainers continued their mode of life, as is usually done in companies maintaining a separate abstainers’ department. Furthermore, this company ruled very strictly in accepting users of alcohol, so that the abstainer is compared with an unusually temperate type of insured ‘risk’.

In 1900, the Equitable Life established an abstainers’ class and admitted new entrants until 1906. People were eligible who had been total abstainers for a number of years. The dividends were effective at the end of ten years, and depended on the mortality in the abstainers’ class. Judging from cases in which dividends have been paid, the death-rate in the abstaining class has been at least 25 per cent less than in the general class.

In the Phoenix Mutual Life Insurance Company, of Hartford, the mortality among abstainers accepted between 1851 and 1861, as compared with mortality in the general group of policyholders accepted since 1861, showed an advantage of 15 per cent among abstainers, notwithstanding the improvement in insurance mortality that has taken place in recent years.

The Security Mutual Life, of Binghamton, New York, established a total-abstainers’ class in 1900. Up to the present the mortality for this class is 30 per cent less than for the general class.

The Manufacturers’ Life of Canada has had an abstainers’ section for the past twenty-nine years. In 1906-1910, the difference in favor of the abstainers was 40 per cent; in 1910-1916, 35 per cent. These gains of the abstainers are over a very favorable general mortality, which tends to strengthen the view that the chief reason for the mortality difference is alcohol.

The Peoria Life has had an abstainers’ section for the past seven years, and reports a difference in favor of abstainers of 25 per cent.

Another important American company gives me the following figures: abstainers show a death-rate 37 per cent lower than moderate, but not daily, drinkers; 32 per cent lower than steady users drinking less than three glasses of beer or two whiskies daily; and 50 per cent lower than those using more than three glasses of beer or two whiskies daily. All members of these drinking classes were accepted as supposedly temperate standard risks.

Not the least important feature of the investigation conducted by the forty-three companies was the mortality figures in occupations where alcohol figured as a hazard.

These were as follows: —

These figures indicate that saloonkeepers have a death-rate higher than that of underground mine foremen; that brewery foremen, maltsters, and the like have a death-rate higher than electric linemen, glass-workers, city firemen (laddermen, pipemen, hosemen), metal grinders or hot-iron workers, although there is nothing in the brewery or saloon business per se that is at all hazardous or unhealthful, aside from the possible temptation to drink and its collateral hazards. Proprietors of distilleries are obviously not so directly exposed to temptation or to other adverse influences that obtain in the retail liquor trade; this accounts for the favorable mortality.

Among hotel-keepers tending bar the death-rate from cirrhosis of the liver was six times the normal; from diabetes, three times the normal; from cerebral hemorrhage or apoplexy, nearly twice the normal; from organic diseases of the heart, nearly twice the normal; from Bright’s disease, nearly three times the normal; from pneumonia, nearly twice the normal.

For brewery officials insuring under 45, the death-rate from cancer and other malignant tumors, cerebral hemorrhage and apoplexy, organic diseases of the heart, pneumonia, and Bright’s disease, among the proprietors, managers, and superintendents is about twice the normal, and from cirrhosis of the liver three times the normal. The death-rate from suicide is nearly twice the normal.

While there are individual differences in various companies, due to the varying standards of selection and classification, nevertheless the general trend of mortality is the same in all companies and shows that ‘Old Mortality’ and ‘John Barleycorn’ are exceedingly good cronies. Wherever you find alcohol you find the following formula at work: More alcohol = higher deathrate.³

III

Nevertheless, scientific caution and veracity impose the obligation not to accept the life-insurance figures at their face value, but carefully to analyze and interpret them in the light of available knowledge regarding the physiological effect of alcohol, and also with due regard to the conditions under which these risks were selected. In judging the matter, all preconceived notions and traditions arising from the fact that alcohol has been closely interwoven with our social fabric for centuries should be excluded. Its age-long use is no proof that it is not a noxious and destructive agent, holding back the human race from higher planes of living. The postulate that time has justified alcohol involves an affirmation that we hold present conditions permanently satisfactory; that crime, immorality and disease have reached an irreducible minimum.

There is, as a matter of fact, no evidence that man is either physically or mentally superior to what he was in the days of ‘the Greeks, our masters.’ On the contrary, there is much evidence that man is physically degenerate, notwithstanding the fall in the total deathrate due to the checking of epidemic disease and infant mortality. Both with and without alcohol nations have risen and fallen. The post hoc ergo propter hoc principle of reasoning is as unsafe in interpreting the effect of alcohol upon life-insurance mortality as its effect upon the historical progress of nations. In considering these things we must avoid the fallacious implication in the expression, ‘ survival of the fittest,’ and remember that many organisms and races survive because they are barely just good enough, not because they have reached perfect adjustment and attained the highest possible development.

In reviewing the evidence offered by the life insurance companies, we must scrutinize the conditions under which the ‘risks’ were accepted. Whatever may have been the earlier ideas regarding total abstinence, it is well known that the non-abstaining policyholders in the British company whose experiences have been quoted were very carefully selected; indeed, the greater confidence in the longevity of abstainers manifested by the establishment of special departments for them no doubt influences the management of such companies to accept slightly impaired abstainers with greater confidence than slightly impaired users; hence there is no reason to believe that the general physical or social type of the abstainers was superior to that of the general class.

In the American companies especially, the custom has been to rule strongly against the alcoholic. One large company with which I was connected for many years had an almost inflexible rule that intoxication within one year of the date of application disqualified a candidate; free drinking, exciting some doubt as to the future, disqualified for a period of at least three years after reform, if not permanently, depending upon the extent of the habit; and treatment for the liquor habit called for rejection. It will be noted that this was wise practice, according to the mortality figures just produced.

This is a small group, and would require close checking as to the homogeneity, except for alcohol, of the drinkers and non-drinkers; but in the light of the larger statistical evidence, it offers interesting confirmation of the fact that high mortality follows the flag of alcohol. — THE AUTHOR.

It is extremely important to bear this in mind, for the reason that these standards of selection show that the mortality figures as to drinkers are derived from supposedly favorable types, and that many individuals in the general population admitting the same degree of indulgence would show a higher mortality. That is, a policyholder admitting an indulgence of three glasses of beer daily would show a lighter mortality than the average man in the population indulging to the same degree. The application of a person suspected of being seriously tainted with liquor is never knowingly accepted on standard forms of insurance by any company. The same principle holds good for persons engaged in those special occupations in which liquor is a hazard. The habits of all such candidates are closely scrutinized, and the benefit of any reasonable doubt is given to the company rather than to the applicant.

It is commonly stated that, inasmuch as these classifications as to the degree of drinking were based on the applicants5 own statements, they cannot be accepted as accurate, and that the excess mortality among the alleged moderate users of alcohol was due to the admission of cases of marked intemperance or to the subsequent development of intemperance among members of that class. Some allowance must of course be made for such a factor, but the broad assumption that the life insurance companies accepted without question the testimony of applicants as to their habits is without warrant.

Even in the British companies, some evidence as to character other than the applicant’s own statement is required; and the medical examiner is also required to pass upon the apparent truthfulness of the statements regarding the use of liquor, past and present.

In American companies, there is a very careful investigation made of the habits of life of all applicants. The ‘moral hazard’ is considered quite as carefully as the physical hazard, and a life that is morally impaired or seriously threatened with moral impairment is rejected. In other words, a man who is in danger of life-failure is not considered a good ‘ risk.’ Liquor is one of the main factors in impairing the moral hazard. A life insurance company looks on this matter from a cold-blooded business standpoint, entirely apart from any standards of social morality. Men are often rejected because of drink who are still socially respected; and men are accepted for life insurance who are not socially respected, provided their habits are not injurious to health and their future seems reasonably secure. There has for years prevailed among life-insurance offices a distrust of the man who drinks every day, even in so-called moderation; and the applications of such persons are always scrutinized very carefully before acceptance. In the main, therefore, the classification of these persons with respect to their consumption of alcohol may be regarded as sufficiently accurate for the purpose of the inquiry.

IV

The question of the relative physical condition of abstainers and of the various classes of insured drinkers having been considered, let us inquire as to the relative social character of the two groups. It has often been claimed that abstainers are a mean lot, often too stingy to die or indulge in anything except long life, and that they are drawn largely from occupations characterized by conservative living, such as that of clergyman, schoolteacher, college professor, and the like. In this connection, a summary of the statements of Mr. Roderick McKenzie Moore, Actuary of the United Kingdom Temperance and General Provident Institution, is entitled to a large measure of confidence. I have the best means of knowing that he is a man governed by scientific caution and veracity, desirous of presenting the facts free from bias, purely as a contribution to science.

He has stated that ‘the total-abstainer class 5 was not nursed, or favored to produce a low mortality. So far as could be determined (and many of the risks came in personal contact with the officers) they were of the same general class as the non-abstainers. They were written by the same group of agents, for the same kind of policies, for the same average amounts, and were in the same general walks of life, and of the same general financial condition. They were almost equal in numbers to the general class and did not form a small high-grade section of the policyholding body. On the contrary, greater care was exercised in the selection of the users of alcohol because of the less favorable experience anticipated with them, and many border-line ‘risks’ were accepted in the abstaining class because of a feeling that their abstinence would neutralize some unfavorable factor.

These statements of the expert, who has personal contact with many of the cases in question, are of greater value than the off-hand criticisms of those who have no first-hand knowledge of the group and who have never taken the trouble carefully to read and digest the evidence.

As to American evidence along similar lines, the analysis of the Security Mutual’s abstaining group also shows that total abstainers are drawn from every class in the community.

There is a widespread impression that total abstainers are greatly in the minority in the population and among insured lives. To throw some light on this subject, two companies, at my request, followed back for a number of months their recent applications, which were in such form that the total abstainers could be distinguished from the users of alcohol. In the Postal Life 64 per cent were abstainers; in the Germania Life 55.8 per cent were abstainers. The rejection rate, excluding those who were rejected on account of intemperance, was about the same in the two classes of applicants, suggesting that total abstainers are in fact only average people, not a small group of ‘health cranks.’

In the New England Mutual an analysis of 180,000 cases insured during the past sixty years showed 24 per cent abstainers and 11 per cent rarely using alcohol, or about 35 per cent practically abstaining from alcohol. The abstainers from tobacco were in about the same proportion. The low mortality among the abstainers in this company has already been mentioned.

The experience of the Life Extension Institute among the individuals it has examined is along similar lines. These examinations were for hygienic or life-lengthening purposes, and included large groups of supposedly healthy, ‘ average ’ people whose employers had subscribed for this service. Among industrial workers 45 per cent were abstainers. Among commercial workers (low average age) 72 per cent were abstainers.

The possible influence of psychopathic or “ nervous” states and of excessive use of alcohol in the non-abstaining group must be considered, as already suggested. But here caution is necessary, and the evidence I have submitted bearing upon the characteristics of these groups and the standards maintained in their selection must be clearly borne in mind. That such conditions — existing psychopathic states and already attained intemperance—were factors of any considerable importance at the time of acceptance, is not to be regarded as a reasonable hypothesis. Psychopathic conditions, including excessive or palpably injurious indulgence in alcohol, developing after the ‘ risks’ had been on the books, must be accepted in the main as a charge against so-called moderate drinking. They are quite as much a possible effect of moderate drinking as any of the many other pathological conditions that are known to result from steady drinking, such as cirrhosis of the liver, fatty liver, or kidney affections, or the various forms of nervous disease or life-failure that may result from the psychic disturbances due to alcohol.

[The second of Dr. Fisk’s papers will deal with Alcohol and Physiology.]