72 
Weights of Human Viscera 
does raise the variability somewhat, i.e. introduces disturbances in the relationship, 
we are not surprised to find that it weakens, although again not in a very marked 
degree, the correlation between weights of heart and spleen. 
(6) On the other hand, Valvular Disease and Aortic Aneurism send up the 
weights of both heart and spleen not only vastly above their healthy values, but 
markedly above the values for the general hospital population. Further, their 
absolute variabilities are increased considerably above the general hospital popu- 
lation values, and d fortiori above the healthy values. The coefficients of 
variation of both are also far above the values in health, and that for the spleen 
above the general hospital values. In the case of the heart, the mean has been 
sent up so high that although the absolute variability is considerably greater, the 
relative variability remains much the same. The general effect of these heart 
diseases is to render the correlation between heart and spleen hardly sensible. 
If we may judge by these two cases the general effect of disease is to increase 
the variability of affected organs and reduce their correlation. This is absolutely 
in keeping with the sensible, but of course less marked, changes we find when we 
pass from a population with healthy organs to the general hospital population, 
which of course contains much disease. 
6. General Conclusions. 
The present paper is chiefly intended as an illustration of how effective 
biometric methods might be from the standpoint of medical science if only there 
were a systematic collection on a large scale of normal and pathological data. 
Some definite conclusions, however, may yet be drawn, and some suggestions made 
on the basis of our numbers. We see sensible, if moderate, correlations between 
the weights of heart, liver, spleen, and kidneys. 
It may be somewhat difficult to understand why the heart-kidney correlation 
is higher than that of the heart with any of the other organs. That the kidney 
should be more closely associated with the heart than the liver is possibly owing 
to its more subordinate functions. The liver is the seat of so many important 
processes that its immediate connection with the heart is not so great as that of 
the heart with the kidney. The excretion of fluids is so closely bound up with 
physical changes in the vascular system, and conversely changes in renal structure 
react so markedly upon the heart and blood vessels that a very close physical 
relationship appears probable. No doubt the liver is greatly affected in many 
forms of cardiac disease, but on the other hand serious functional disturbances or 
even acute inflammation of the liver do not produce heart changes with the same 
precision as analogous mischief in the kidney does. If this partially and imperfectly 
accounts for the higher heart-kidney correlation as compared with the heart-liver 
coefficient, it may perhaps serve as an argument in the case of the spleen. Heart 
mischief nearly always reacts on the spleen, but splenic trouble does not always 
affect the heart. 
