64 
This question can be answered only approximately. At the Con- 
necticut hospital we found a difference of about 9 per cent between 
the amount of infection present among males and among females; about 
10 per cent may be due to a difference in age, about 5 per cent to a 
difference in the length of institutional life, and about 2 per cent to 
nativity. The proper combination of these conditions, therefore, 
might be held accountable, perhaps, for a maximum difference of 
from 20 to 30 per cent in the frequency of infection upon the different 
wards and sections. The range of the rate of infection upon the differ- 
ent wards is from 0 to 75 per cent and upon the different sections 
from 2.56 to 61.33 per cent. There remains, therefore, from 30 to 45 
per cent of variation in the amount of infection upon the different 
wards and sections which could not be explained by any combination 
of sex, age, nativity, or length of hospital life. 
As we find these factors in only a very imperfect combination upon 
our wards, it is probable that they can be held accountable in oniy a 
small measure for the unequal distribution of infection among the 
wards. Since no conditions, sanitary or administrative, were found 
which would account for more infection upon one ward than upon 
another, and since the unequal distribution of infections has not been 
adequately explained by conditions of sex, age, length of institutional 
life, or nativity, we must content ourselves with the mere presentation 
of the facts as they appear. 
RESULTS OF FORMER INVESTIGATIONS. 
In Table 3 we have brought together most of the statistics at our 
command concerning the frequency of intestinal worms in man. It is 
difficult to draw any comparisons between these different results. 
The very high percentage of infection found in some investigations is 
due to an endemic condition with a certain worm in the locality in 
question, in others to a selection of cases. It is notable, however, 
that our own results show the lowest total rate (11.17 per 100) of 
infection, the 14.07 per 100 reported by Mueller from 1,939 autopsies 
made at Dresden being the next lowest. From these figures the rate 
of infection varies through almost all degrees until it reaches its 
maximum (139.65 per 100) among the 4,482 anemies examined by 
Ashford, King, and Gutierrez in Porto Rico. 
There is also a marked variation in the relative frequency of the 
different worms as given by the different authors. A glance at the 
table shows that the three nematodes, Trichuris , Ascaris, and Oxyuris, 
give the great majority of the infections. The hookworm is the most 
common parasite in the Porto Rican statistics (100 per cent), in those 
of Dobson (75.78 per cent) and of Fearnside (65.83 per cent) in India, 
and of Daniels in British Central Africa (9.96 per cent). Boycott 
(4.08 per cent) and Strong (2.48 per cent) are the only additional 
authors besides ourselves (1.04 per cent) who give data for this worm. 
