126 
PA THOLOG Y: PEARL A ND BA CON Proc. N. A. S. 
give a picture of regulatory processes perhaps more detailed and definite 
than is likely to be found elsewhere, it seemed probable that quantitative 
study of the ratios existing between the weights of the more important 
viscera would indicate with great exactness any disturbances in the normal 
balance of the parts of the organism. The results, we believe, fully bear 
out this assumption. This first investigation, which has just been com- 
pleted, was restricted entirely to tuberculosis as the pathological condition 
involved. This choice was deliberate and based upon the known pro- 
found constitutional effects of the disease. The data studied are six 
organ-weight indices or ratios, defined as follows : 
Iviver weight 
Index A = ^— = L/H 
Heart weight 
Liver weight 
Index B = — = L/S 
Spleen weight 
Liver weight 
Index C = - L/K 
Right kidney weight + Left kidney weight 
Heart weight 
Index D = — = H/S 
Spleen weight 
Right kidnev weight + Left kidney weight 
Index K = — = K/H 
Heart weight 
Right kidney weight + Left kidney weight 
Index F = — — = K/S 
Spleen weight 
These indices were calculated from the data given on the protocols of 
such of the first 5000 autopsies (1341 in all) of the Johns Hopkins Hos- 
pital as revealed tuberculous lesions, and at the same time conformed to 
certain other requirements. The material was divided for purposes of 
study by race, sex, activity or quiescence of lesion at time of death, age, 
and character of lesion. 
The age characteristics of this autopsied, tuberculous, hospital population 
were compared in detail with those of the general population, and with that 
part of it dying of tuberculosis. In general there is a close agreement in 
mean age at death between the autopsied hospital population showing 
active tuberculous lesions, and the portion of the general population 
dying of tuberculosis. 
One of the most interesting and novel results of the study has been to 
show that certain of these organ-weight indices change with the age of the 
individual. These changes may be looked upon as phenomena of senescence 
beginning very early in life, indeed practically at birth. The detailed re- 
sults as to correlation with age may be briefly summarized as follows. 
There is a significant negative correlation in all groups between the value of 
Index A and age. This means that as age advances the liver /heart 
ratio tends to take on lower and lower values. Index B is, on the whole 
