Ethel M. Elderton and Karl Pearson 
567 
which seems to suggest that other things being constant increasing incidence is to 
some slight extent followed — probably as the only suggested remedy — by the 
higher isolation rates*. 
(9) Can any other Factors he determined which measure the Relation between 
urban conditions and the Incidence of Diphtheria ? It is worth while from this 
standpoint to place the towns with which we have dealt in the order of incidence, 
each town being credited with the mean of the three attack -rates for each of 
three-year periods. Now an examination of the four columns of this table shows 
that, with the exception of Oxford — which has a child incidence ('89 as com- 
pared to '70) considerably above the population incidence owing to relatively 
few children — the towns with the least diphtheria are the Midland, and parti- 
cularly the Northern manufacturing towns. These constitute practically the 
whole of the first column of 19 towns. The last column contains the big ports 
and certain suburban metropolitan districts, indeed all these for which we have 
data except Plymouth, Devonport and Tottenham fall into the second half of the 
Seventy-six Towns in order of their Diphtheria Incidence Rates 1904-1912. 
1 
West Bromwich 
(•40) 
20 
Rotherham 
(•91) 
39 
Birkenhead 
(V20) 
58 
Brighton 
(1-61) 
2 
Northampton 
(•45) 
21 
South Shields 
(•92) 
40 
Rhondda 
(r24) 
59 
Stockton 
(1-64) 
3 
Wigan 
(•48) 
22 
Preston 
(•93) 
41 
Smethwick 
(1-25) 
60 
Grimsby 
(1^66) 
4 
Walsall 
(•49) 
23 
Wallasey 
(•94) 
42 
Barrow 
(r25) 
61 
Leyton 
(1^73) 
5 
Stockport 
(•53) 
24 
Bath 
(•95) 
43 
Newport 
(1-25) 
62 
West Ham 
(1-75) 
6 
Oldham 
(•59) 
25 
Bootle 
(•96) 
44 
Wimbledon 
(1'30) 
63 
Salford 
(1^78) 
7 
Bolton 
(•59) 
26 
York 
(•99) 
45 
Great Yarmouth 
(1'31) 
64 
Nottingham 
(V79) 
8 
Oxford 
(•70) 
27 
Blackpool 
(•99) 
46 
Southend -on-Sea 
(1-32) 
65 
St Helens 
(1-80) 
9 
Barnsley 
(•71) 
28 
Tyncmouth 
(1-00) 
47 
Birmingham 
(1-32) 
66 
Walthamstow 
(1-83) 
10 
Southport 
(■72) 
29 
Tottenham 
(1-02) 
48 
Gillingliam 
Ipswich 
(1-34) 
67 
Ilford 
(1-87) 
11 
Rochdale 
(•73) 
30 
Halifax 
(1^03) 
49 
(1-36) 
68 
Southampton 
(1-95) 
(2^03) 
12 
Leicester 
(-76) 
31 
Sheffield 
(1'03) 
50 
Liverpool 
(1-37) 
69 
Cardiff 
13 
Manchester 
(•79) 
32 
Plymouth 
(1-07) 
51 
Hornsey 
(1-39) 
70 
Enfield 
(2-11) 
14 
Bury 
(•80) 
33 
Coventry 
(Ml) 
52 
Darlington 
(1"39) 
71 
Hull 
C2^16) 
15 
Blackburn 
(•83) 
34 
Warrington 
Devonport 
(I'll) 
53 
Acton 
(1-43) 
72 
Bristol 
(2-20) 
16 
Wolverhampton 
(•86) 
35 
(ri3) 
54 
Newcastle 
(1-45) 
73 
Croydon 
(2^41) 
17 
Burnley 
(•86) 
36 
Sunderland 
(M4) 
55 
Burton on Trent 
(1^48) 
74 
Portsmouth 
(2-51) 
18 
Huddersfield 
(•89) 
37 
Bournemouth 
(1"17) 
56 
Bradford 
(1^56) 
75 
Derby 
(3-15) 
19 
Wakefield 
(■91) 
38 
Middlesbrough 
(r20) 
57 
Willesden 
(r56) 
76 
Lincoln 
(3^65) 
* It is perhaps worth while putting on record the additional statistical constants obtained in 
deducing the above correlations, as they are probably fairly reliable values and should be compared 
with the two period constants on p. 561 : 
A' = Mean Attack-rate 1^26 ; Standard Deviation, Attack-rate -655 
i' = Mean Isolation-rate 47 75; Standard Deviation, Isolation-rate 26-341 
5j^' = Mean Increase in Attack-rate - -080 ; Standard Deviation of change in Attack-rate -648 
5i/J' = Mean Increase in Isolation-rate 9-03 ; Standard Deviation of Increase in Isolation-rate 1-05 
Thus while most towns have been sensibly increasing their amount of isolation by 17 °/o to 18 % of 
its mean value, the decrease in the attack -rate has only been 6% to 7% of the mean incidence, and 
the correlations show that this decrease has not occurred in the towns with marked increase of 
isolation. 
