Value of 
Family History and Personal Condition 
IN 
ESTIMATING A LIABILITY 
TO 
CONSUMPTION 
Published by 
The Mutual Life Insurance Company 
OF NEW YORK 
1895 w ith the Compliments of 
The Mutual Life Insurance Company of New York 
s President. VALUE OF 
Family History and Personal Condition 
IN 
ESTIMATING A LIABILITY 
TO 
CONSUMPTION 
Published by 
The Mutual Life Insurance Company 
OF NEW YORK 
1895 Copyrighted by 
The Mutual Life Insurance Company of New York. 
November, 1895. Value of Family History and Personal 
Condition in Estimating a Liability to 
CONSUMPTION 
Richard A. McCurdy, Esq., 
President. 
Sir: 
About twenty years ago a complete analysis was 
made of the Mortuary Records of this Company up to 
that date. Especial attention was given to the losses 
from consumption, and some valuable conclusions were 
drawn from the results of this experience. Since that 
time, similar investigations have been pursued, and statis- 
tics have been compiled by physicians from the experience 
of other Life Insurance companies and from general 
medical practice. The main subjects of these studies 
have had reference to the transmission by inheritance of 
consumptive tendencies, and the respective values of 
family and personal qualities as indicating a liability to 
consumption. Very recent discussions in the medical 4 
societies show that the questions are still unsettled, and 
that a wide latitude of opinion prevails among physicians. 
Believing that the experience which has been acquired 
by this Company since the former report might tend to 
settle some of these undetermined problems, I have care- 
fully examined and analyzed our recent records, and 
submit the following report. 
The material used for this report consists of 
two thousand and eighty-five cases, being the entire mor- 
tality during the fifteen years from 1879 to 1893. The 
original application and death proofs were examined, and 
an abstract made on cards of the family record and per- 
sonal condition at the time of insurance, and the cause of 
death. In a very few of the earliest cases the statements 
are deficient on some of these points. From this material 
all the cards were selected which gave any record of con- 
sumption in the family. The number of such cases was 
nineteen hundred and ninety-four. A class of cases with 
“ tainted record ” is thus made, from which the importance 
of such taint can be deduced. The class thus separated 
might be compared with the “total remaining,” but for 
satisfactory reasons it is preferable to employ only an 
approximately equal number of cases. This has been 
done by taking all the deaths which occurred in the 
beginning, middle and end of the whole period, namely, 
in 1879, 1886 and part of 1893, amounting to twenty-seven 
hundred and six cases. Two classes are thus made, which 5 
differ in no respect except as to the presence or absence 
of a family record of consumption at the time of insurance. 
Proceeding to an analysis of this experience, we have: 
1 ABLE No. I. 
SHOWING THE MORTALITY FROM CONSUMPTION IN NON- 
CONSUMPTIVE AND CONSUMPTIVE FAMILIES. 
Age at Death and Cause 
of Death 
Non-consump- 
tive Families 
Consumptive 
Families 
All causes 
106 
60 
To 29 years h 
Consumption 
38 
3i 
Percentage 
35-S 
51.6 
All causes 
372 
194 
30 to 39 “ H 
Consumption 
98 
77 
Percentage 
26.3 
39-6 
All causes 
566 
284 
40 “ 49 “ -j 
Consumption 
100 
70 
Percentage 
17.6 
24.6 
All causes 
724 
489 
50 “ 59 “ 
Consumption 
49 
77 
Percentage 
6.7 
L5- 7 
All causes 
613 
569 
60 “ 69 “ h 
Consumption 
36 
47 
Percentage 
5-8 
8. 2 
All causes 
325 
398 
70 “ 79 “ - 
Consumption 
12 
8 
Percentage 
3-7 
2.2 
Total Deaths 
2706 
1994 
This table shows that the percentage of deaths from 
consumption has been much larger in the “ tainted” than 6 
in the “ non-tainted ” class; and that this increased pro- 
portion exists at every decennial period of life until 
extreme old age. 
To form a correct estimate of the value of such 
record, as an indication of the liability to consumption, 
the cases should be viewed as they present themselves 
when applicants for insurance. It is at this time that 
the medical examiner must decide as to the eligibility 
of the proponent, and the family record is considered an 
important element in this decision. The age of the 
applicant himself may be a qualifying condition, for 
the influence of family inheritances and tendencies is 
believed to vary at different periods of life. 
In the following table, therefore, the cases have been 
grouped according to the age at insurance: 
Table No. 2. 
SHOWING THE MORTALITY FROM CONSUMPTION IN NON- 
CONSUMPTIVE AND CONSUMPTIVE FAMILIES. 
Age at Insurance and Cause 
of Death 
Non-consump- 
tive Families 
Consumptive 
F amilies 
All causes 
4 73 
255 
To 29 years 
Consumption 
135 
96 
Percentage 
28.5 
3 7 A 
( All causes 
9°3 
674 
30 to 39 “ 
Consumption 
120 
136 
. Percentage 
13- 3 
2.0.2 
’ All causes 
8*3 
596 
40 " 49 “ < 
Consumption 
54 
55 
Percentage 
6.8 
y. 2 
50 years and 
upwards 
' All causes 
Consumption 
Percentage 
517 
24 
4.6 
469 
2 3 
4.0 7 
It would appear from this table that the importance 
which has been attached to a family record of consump- 
tion is not modified by the age of the applicant; for the 
influence of this taint is shown as clearly in persons 
between forty and fifty years as in those between twenty 
and thirty years of age. With regard to those who were 
above fifty years at the time of insurance, there has been 
no difference between the two classes of “tainted” and 
“non-tainted” in rate of mortality from consumption; 
but, the actual number of deaths from this disease having 
been very small, the natural inference is not conclusive. 
The higher rate of difference between the two classes at 
the ages of thirty to forty years is also probably accidental, 
and might be modified in another series of observations. 
It must be remembered that these tables indicate only the 
rate of consumption in the total mortality, and not the 
number of deaths that have occurred in proportion to the 
number of living at each period of life. The actual mor- 
tality from consumption remains nearly constant at all 
ages, but its ratio among the total causes of death 
diminishes with advancing years, on account of the 
appearance and rapid increase of other diseases. In early 
manhood, consumption is a principal cause of death; after 
middle age, degenerative diseases of the circulatory and 
nervous systems cause it to assume a less conspicuous 
position. 
In the table given on page 6, the percentage of con- 8 
sumption for persons insured at ages twenty to thirty 
years was 28.5 per cent, among those with untainted 
record, and about one-third greater, or 37.6 per cent., 
among those with tainted record; at ages from forty to 
fifty it was 6.8 per cent, for the untainted, and one- 
third greater, or 9.2 per cent., for the tainted; the pro- 
portion of increase being the same in both cases. From 
these figures it is evident that a consumptive famity 
record is of considerable value as indicating an increased 
liability to consumption in an applicant, and that the 
amount of this increased liability may be estimated at 
thirty per cent. This estimate holds good until the age 
of fifty years is reached, after which age no difference is 
apparent between the classes. For the latter period, 
however, the statistics are too small to warrant any safe 
conclusion. 
The degree of inherited predisposition, and the 
resulting liability to a disease, is often estimated by the 
number and relationship of the members of the family 
who have been previously affected. A family history 
exhibiting two or more consumptives is generally con- 
sidered more significant than the record of a single 
case; and the death of a parent as more ominous 
than that of a brother or sister. 
Life Insurance experience and statistics cannot 
settle this point, but may throw some light upon it. 
The original selection of the risks was influenced by 9 
the family record of the applicants, and applicants 
belonging to families in which the consumptive taint 
was strongly marked were excluded from insurance. 
A record of the disease in both parents, or in two or three 
members of the family, was, as a general rule, considered 
a bar to acceptance, and therefore our experience in this 
class of cases is comparatively limited. Almost all of 
those admitted with a record showing taint in two 
members were above forty years of age, partly as the 
result of unfavorable selection at younger ages, and 
partly for the natural reason that as life advanced the 
family record would be more complete, and would give 
more deaths from all causes, including consumption. 
An elaborate table has been prepared, and is here 
given (Table No. 3), showing the proportionate mortality 
from consumption and other causes among those with 
tainted record, and arranged according to the degree of 
taint and the age at insurance. 10 
Table No. 3. 
SHOWING THE PROPORTIONATE MORTALITY FROM CONSUMPTION AND OTHER 
CAUSES AMONG THOSE WITH TAINTED RECORD, AND ARRANGED ACCORD- 
ING TO THE DEGREE OF TAINT AND THE AGE AT INSURANCE. 
Family Record of Consumption 
Age at Insurance 
Member of Family Affected 
Total 
20-29 
Years 
30-39 
Years 
40-49 
Years 
SoYears 
Up- 
wards 
No member 
affected 
( Deaths from all causes 
■< “ Consumption 
( Percentage . . 
2706 
333 
12.7 
473 
*35 
28.5 
9°3 
120 
13-3 
813 
54 
6.8 
5*7 
24 
4.6 
Both parents 
( Deaths from all causes 
} ‘ ‘ Consumption 
[ Percentage . . . 
7 
1 
14-3 
1 
2 
1 
50.0 
4 
Father . . 
' Deaths from all causes 
‘ ‘ Consumption 
' Percentage .... 
343 
66 
20.7 
69 
3° 
43-3 
136 
22 
16.1 
86 
9 
10.4 
52 
5 
9.6 
Mother . . 
' Deaths from all causes 
“ Consumption 
' Percentage .... 
457 
7i 
i5-5 
79 
29 
36.5 
156 
28 
17.1 
135 
10 
7-5 
87 
4 
4.6 
Parent and 
brother or 
sister . . 
' Deaths from all causes 
“ Consumption 
’ Percentage .... 
120 
13 
11.0 
11 
2 
18.2 
42 
9 
21.4 
36 
2 
5-5 
3i 
One brother 
or sister . 
' Deaths from all causes 
“ Consumption 
1 Percentage .... 
933 
133 
M-3 
88 
30 
34-1 
306 
63 
20.6 
284 
29 
10.2 
255 
11 
4-3 
Two or more 
brothers or 
sisters . . 
' Deaths from all causes 
“ Consumption 
1 Percentage .... 
134 
26 
19.4 
8 
5 
62.5 
33 
14 
42.4 
53 
4 
7-5 
40 
3 
7-5 
Two m e m - 
bers affect- 
ed . . 
Deaths from all causes 
“ Consumption 
Percentage .... 
261 
40 
15-3 
19 
3^-8 
76 
23 
3°-3 
9i 
7 
7-7 
75 
3 
4.0 
One member 
affected 
Deaths from all causes 
“ Consumption 
Percentage .... 
1733 
270 
15.6 
236 
89 
37-2 
598 
”3 
18.5 
5°5 
48 
9.2 
394 
20 
5-i 
Parent and 
brother or• 
sister . . i 
Deaths from all causes 
“ Consumption 
Percentage .... 
120 
*3 
11.0 
11 
2 
18.2 
42 
9 
21.4 
36 
2 
5-5 
31 
Two or more i 
brothers or - 
sisters . . 1 
Deaths from all causes 
“ Consumption 
Percentage .... 
134 
26 
19.4 
8 
62.5 
33 
14 
42.4 
53 
4 
7-5 
40 
3 
7.5 
One parent . - 
Deaths from all causes 
“ Consumption 
Percentage .... 
800 
137 
17.1 
148 
59 0 
39-8 
292 
5° 
17.1 
221 
19 
8.6 
*39 
9 
6-5 
One brother \ 
or sister j 
Deaths from all causes 
‘ ‘ Consumption 
Percentage . . . . 
933 
133 
14.3 
88 
3° 
34.1 
306 
63 
20.6 
284 
29 
10.2 
255 
11 
43 11 
Under this tabulation the numbers for each sub- 
division are small, and the fluctuations great, and 
consequently the deductions or conclusions should be very 
guarded. On one point, however, the evidence is 
conclusive, namely, that the death of a brother or sister 
is of equal significance with that of a parent. 
There has been little restriction in the acceptance of 
risks in which the family history showed only a single 
death from consumption, whether this death may have 
been in a parent or brother or sister; consequently, the 
experience with this degree of taint has been large and 
correspondingly reliable. A comparison of the two 
subdivisions of family taint—taint in a parent, and 
taint in a brother or sister—shows decided fluctuations 
in the mortality from consumption, sometimes one class 
and sometimes another presenting the highest percentage; 
but, on the whole, the results are nearly equal. The pre- 
disposition or tendency to consumption is rather a 
“ family ” than an “ inherited ” influence. It may pro- 
ceed directly from a phthisical or other constitutional 
taint in the parents, or may be acquired after birth by the 
conditions and environment common to all the members 
of the family during the periods of childbirth and youth, 
including the possibility of direct personal contagion. 
The significance of this kind of “ family ” taint is shown 
with still greater force in the subdivision marked “two or 
more brothers or sisters.” In this group the percentage of 12 
deaths from consumption reaches the maximum, and at 
the earlier ages far exceeds that of any other group. The 
death of both parents has been considered to indicate such 
an extreme danger of taint in the descendants, as to be 
generally a bar to insurance. Consequently, we find in 
our records only seven of this class, who were probably 
accepted as exceptions to the general rule on account of 
extremely favorable conditions of age, physique and 
habits. The appearance of consumption in two genera- 
tions is also considered very unfavorable, but, probably 
owing to extreme care taken in the examination and 
selection of applicants with this history, our experience 
has been very favorable. The subdivision showing a 
record of “parent and brother or sister ” gives the very 
smallest percentage of consumption of any class with 
tainted family record, equal in fact to the general class 
with no taint. The total number in this subdivison is not 
large, and the results therefore may be accidental; but I 
believe that special care in selection, in view of the recog- 
nized danger of family predisposition, has had a decided 
influence in effecting the favorable results. 
This leads naturally to the second point for investiga- 
tions, namely, whether favorable personal conditions may 
not equalize or even overbalance unfavorable conditions 
of family predisposition. The record of a death from 
consumption establishes a presumption that the other 
members of the family have a predisposition to, and an 13 
increased probability that some of them will die from this 
disease. Only a small proportion of these individuals, 
however, subsequently do die of consumption, and there- 
fore there must be some other conditions which neutralize 
or counteract this tendency. It is probable that one of 
these personal conditions is that which is generally known 
as robustness, indicated by muscular vigor and strength, 
weight of body and capacity of chest. A prerequisite for 
this result is a condition of the digestive processes, shown 
by the power to take up and assimilate a large amount of 
nutriment. Our previous examination and report on this 
subject, which has been corroborated by similar reports 
of others, proved that the vast majority of consumptives 
were persons below the average weight; but the statistics 
were not sufficiently numerous to admit of subdivisions 
according to age. Now, the highest percentage of con- 
sumption mortality and the highest proportion of under- 
weights are found before middle age; the least percentage 
of consumption mortality and the highest proportion of 
overweights, after middle age: consequently, as the bodily 
weight and increasing years advance together, there was a 
probability that what was attributed to weight was merely 
due to age. Our present experience should be sufficiently 
large to establish the value of this personal condition in 
indicating a consumptive tendency, or in modifying for 
better or worse any influence of inheritance, and I have 
examined the question by analyzing the series of cases 14 
which is the subject of this report. The total number is 
somewhat reduced, for the reason that in the earliest 
records of the Company the examination papers do not 
show the height and weight. The standard of measure- 
ment mentioned is that habitually used by the Company’s 
Examiners, except that for persons under the age of thirty 
years a deduction of five pounds was made, to correspond 
with the normal weight for that period of life. 
The following table is made from the class in which 
there was no history of family phthisis: 
Table No. 4. 
NO HISTORY OF FAMILY PHTHISIS. 
Age at Insurance and Cause 
Weight 
of Death 
At or above 
Average 
Below 
Average 
To 29 years - 
All causes 
Consumption 
Percentage 
197 
35 
17.8 
226 
93 
41.1 
30 to 39 “ - 
All causes 
Consumption 
Percentage 
412 
35 
8-5 
382 
73 
19.1 
40 “ 49 “ - 
All causes 
Consumption 
Percentage 
476 
22 
4.6 
271 
29 
10.7 
50 years and 1 
upwards 
All causes 
Consumption 
Percentage 
335 
9 
2.7 
151 
14 
9-3 
It is here seen that for persons with untainted record 
there have been twice as many deaths among those 
below than those above the average weight, and that 
this disproportion prevails equally at all ages. Those 15 
below the average appear to be twice as susceptible to the 
disease, and therefore the condition of bodily develop- 
ment is of very great importance in indicating the 
presence of the phthisical predisposition. This indi- 
cation is of equal value at all periods of life; for, while 
the actual percentage of deaths from consumption is 
greatest in early life, the disproportion between the two 
classes continues constant for all ages. On the other 
hand, a good and robust physical development with 
absence of family taint does not ensure against the 
danger of consumption, but only diminishes very greatly 
the probability of its occurrence. 
The next table gives the cases in which there was 
some record of family phthisis. 
Table No. 5. 
SOME RECORD OF FAMILY PHTHISIS 
Age at Insurance and Cause 
Weight 
of Death 
At or above 
Below 
Average 
Average 
All causes 
88 
141 
To 29 years •< 
Consumption 
22 
69 
Percentage 
25.0 
48.9 
All causes 
273 
308 
30 to 39 “ •< 
Consumption 
3i 
87 
Percentage 
IJ*3 
28.2 
All causes 
328 
191 
40 “ 49 “ s 
Consumption 
22 
27 
Percentage 
6.7 
14.1 
50 years and 
upwards 
All causes 
288 
151 
Consumption 
Percentage 
10 
3-5 
11 
7-3 16 
As was to have been anticipated, the percentage 
from consumption is here somewhat greater under each 
subdivision; but the difference between the two classes of 
over and under weight is almost exactly the same as 
in the last table. Those who possessed a physical 
robustness and weight up to the average gave only one 
half as many deaths as those below the average, and this 
proportion was constant at all ages. A comparison of 
figures in the tables will also show clearly that, as far as 
our experience has gone, the family history has influenced 
the consumption death rate far less than has the personal 
condition, and that therefore it is a less valuable indi- 
cation of an underlying phthisical predisposition. A 
proper discrimination of the relative values of these two 
factors is very important, and the following diagrams 
have been prepared with a view to making them more 
clear and striking. It must be remembered, however, 
that the diagrams are based on the rate of consumption 
in proportion to other causes of death, and not on the 
number of deaths in relation to the lives exposed. If 
this were overlooked, it might be inferred that con- 
sumption was of rare occurrence after middle age, whereas 
it is as frequent then as in early life, but its percentage 
is diminished by the steady increase of other fatal 
diseases. 17 
DIAGRAM ILLUSTRATING THE INFLUENCE OF FAMILY RECORD AND PERSONAL WEIGHT, SEPARATELY AND COMBINED, 
IN MODIFYING THE PROPORTION OF CONSUMPTIVE DEATHS IN THE TOTAL MORTALITY. 
Tainted record . |§: ■■ Consumption 48.9 All other causes 51.1 
Below average . Consumption 44.1 All other causes 55.9 
Jntainted record Consumption 41.i All other causes 58.9 
[Tainted . . . J Consumption 37.6 All other causes 62.4 
Untainted . . Consumption 28.5 All other causes 71.5 
Tainted record . : Consumption 25.0 All other causes 75.0 
AGES AT INSURANCE UNDER 30 YEARS. 
Above average . r Consumption 20.0 All other causes 80.0 
Untainted record Consumption 17.8 All other causes 82.2 
Family 
record 
Weight 
Above 
average 
weight 
Below 
average 
weight 18 
DIAGRAM ILLUSTRATING THE INFLUENCE OF FAMILY RECORD AND PERSONAL WEIGHT, SEPARATELY AND COMBINED, 
IN MODIFYING THE PROPORTION OF CONSUMPTIVE DEATHS IN THE TOTAL MORTALITY. 
Untainted . . Iflfc, 13-3 All other causes 86.7 
Tainted record . I Consumption 28.2 All other causes 71.8 
Tainted Consumption 20.2 All other causes 79.8 
Below average . ] Consumption 23.2 All other causes 76.8 
AGES AT INSURANCE 30 TO 39 YEARS. 
Above average . jjP All other causes 90.4 
f Untainted record j 8.5 All other causes 91.5 
Tainted record |$ n-3 All other causes 88.7 
Untainted record j Consumption 19.1 All other causes 80.9 
(Consumption) 
IConsumotion') 
(Consumption) 
(Consumption) 
Family 
record ‘ 
Weight 1 
Above 
average . 
weight 
Below 
average 
weight 19 
DIAGRAM ILLUSTRATING THE INFLUENCE OF FAMILY RECORD AND PERSONAL WEIGHT, SEPARATELY AND COMBINED, 
IN MODIFYING THE PROPORTION OF CONSUMPTIVE DEATHS IN THE TOTAL MORTALITY. 
AGES AT INSURANCE 40 TO 49 YEARS. 
Below average . 12.1 All other causes 87.9 
Tainted record Consumption 16.8 All other causes 83.2 
Untainted . . | 6.8 All other causes 93.2 
Tainted . . .1 9^2All other causes 90.8 
Above average .| 5.5 Allother causes 94.5 
Tainted record . 6.7 All other causes 93.3 
Untainted record 1 10.7 All other causes 80.3 
Untainted record 4.6 All other causes 95.4 
(Consumption) 
(Consumption) 
(Consumption) 
(Consumption) 
(Consumption) 
(Consumption) 
(Consumption) 
Family 
record 
Weight 
Above 
average 
weight 
Below 
average 
weight 20 
DIAGRAM ILLUSTRATING THE INFLUENCE OF FAMILY RECORD AND PERSONAL WEIGHT, SEPARATELY AND COMBINED, 
IN MODIFYING THE PROPORTION CF CONSUMPTIVE DEATHS IN THE TOTAL MORTALITY. 
AGES AT INSURANCE 50 YEARS AND ABOVE. 
Below average . | 8.3 All other causes 91.7 
Untainted record]2.7 All other causes 97.3 
Untainted record f A11 other causes 90.7 
Untainted . .4-6 All other causes 95.4 
I Tainted . . . | 4-9 _ All other causes 95.1 
Above average . 3.0 All other causes 97.0 
Tainted record . 3-5 All other causes 96.5 
Tainted record . I 7-3 All other causes 92.7 
(Consumption) 
(Consumption) 
(Consumption) 
(Consumption) 
(Consumption) 
(Consumption) 
(Consumption) 
(Consumption) 
Above 
average 
weight 
Below 
average 
weight 
Family 
record 
W eight 21 
These diagrams show through the eye what has 
already been demonstrated by the tables of figures, 
namely, that both the history of phthisis in the family 
and the condition of personal underweight indicate an 
increased liability to consumption, and that these factors 
differ considerably in the degree of their influence. The 
family history exhibits much less influence than the 
weight, and a good record of weight overbalances tainted 
family history. The combination of the two factors 
indicates the maximum, and the absence of both, the 
minimum of susceptibility. 
While mindful that individuals with a very unfavor- 
able family record have as a rule been rejected, and that 
therefore our experience has been to this extent limited, 
I believe that the statistics given in this report warrant 
the following conclusions: 
ist. That the history of consumption in any 
member of the immediate family increases the prob- 
ability of its appearance in an applicant. 
2d. That consumption in a brother or sister is 
at least of equal importance as when it has occurred 
in a parent. 
3d. That persons who are under the standard or 
average of weight are much more liable to consumption 
than those above this standard. That the peculiarity of 
constitution which is indicated by the inability to take 
and assimilate a proper amount of nutriment indicates a 22 
susceptibility to phthisis, or at least is a reasonable 
suspicion of such predisposition. 
4th. That persons who exhibit a robust and well 
developed body have little susceptibility to consumption. 
5th. That the personal condition of weight and 
robustness has far more value than the family history in 
diminishing the liability to consumption; therefore, 
6th. The evidence presented by a well developed 
body may outweigh the suspicion attached to unfavor- 
able family record. 
7th. That these influences of family history and 
personal weight are of the same grade for every age, and 
their importance is not lessened by the fact that the 
individual has reached middle life. 
In deciding upon the eligibility of an applicant for 
Life Insurance, in whose case there is a suspicion of 
future danger from consumption, his personal condition 
is of the first, and his family record of the second 
importance. Whenever he presents a robust physical 
appearance, with a weight at least equal to the standard 
or average as given in our tables, he may be accepted, 
notwithstanding any taint in the record of his family. 
In our experience such persons have a small liability 
to consumption, although not protected from it. If, 
however, his weight does not come up to the average 
and he gives a history of consumption in brothers, 
sisters or parents, he is to be regarded as an unfavor- 23 
able risk. This does not mean that all such persons are 
to be absolutely excluded from insurance, but each case 
must be carefully scrutinized, and the decision based 
upon the circumstances of occupation, character, past 
history, etc. When these are favorable, insurance 
should be given on terms most advantageous to the 
Company, by limiting the amount or modifying the 
form of policy; when unfavorable, the applicant should 
be either postponed until he has gained sufficient weight, 
or else be absolutely rejected. 
Respectfully submitted, 
E. J. Marsh, M. D., 
Medical Director. 
New York, April 15, 1895.