394 
this date the disease had suddenly and notably increased, 
a large proportion of the recorded cases having occurred 
amongst the customers of milk-retailers dealing in the 
particular Hendon milk. These facts strengthened the 
case against the Hendon farm, but did not by any means 
establish it, inasmuch as the retailers in question obtained 
their supplies from other farms as well, and although in 
two cases these were situated in widely different counties, 
yet the case against Hendon was still in the condition of 
not proven, more especially as the St. John’s Wood 
customers of that farm were certainly wholly free from 
scarlatina. Simultaneously with this investigation, another 
was being pursued at the incriminated dairy itself. But 
nothing was revealed here to show how the disease could 
be propagated from it as a centre. There was no scarla- 
tina, nor any illness at all like scarlatina, amongst the 
persons employed about the farm, or their families and 
neighbours, at any such time or in any such way as to 
influence the farm or its produce. This, then, was the 
state of affairs on December 23, or less than a week after 
Mr. Blyth’s report: there was a strong presumption 
against the Hendon farm, but outside human agencies 
had to be set aside as not having been operative. A 
thorough inspection of the farm itself was at once under- 
taken. Now it happened that the farmer in question, as 
well as one of the dealers who purchased from him, was 
particularly careful in all sanitary matters respecting his 
dairy. Every precaution had been taken by both to 
secure the farm and milk against any known sanitary 
fault or misadventure, and thus the inquiry advanced 
another stage. If the Hendon farm had caused the 
scarlatina, it did not do so in any commonly accepted 
way, such as through unwholesome conditions of water 
or drainage, or careless handling of milk or milk-utensils, 
by persons carrying scarlatina infection. This threw Mr. 
Power back on the theory of something in the cows 
themselves which caused the scarlatina to be distributed 
with their milk, and this formed his working hypothesis 
thenceforth. To discover this “something,” and to 
understand its nature, it was necessary to ascertain in 
detail every parallel between the doings at the dairy farm 
and the observed scarlatina. 
Here, then, we enter on the second and by far the 
most difficult stage of the investigation. The various dis- 
tricts supplied from Hendon were taken one by one; the 
quantities of milk obtained from Hendon by the dealers 
there, and by the same dealers from other sources, were 
ascertained ; the dates of the notable incidence of the 
disease among the customers, and the degree of incidence 
at one period and another, were carefully observed, and 
compared, with the following results :—(1) The disease 
commenced at one and the same time in the four districts 
supplied from Hendon, viz. South Marylebone, Hamp- 
stead, St. Pancras, and Hendon. (2) In South Mary- 
lebone the disease increased day by day with increasing 
force up to the date of the inquiry. (3) In Hampstead 
and St. Pancras there was a cessation of ten days after 
the first attack, and then a larger number of persons were 
taken ill, the attacks continuing up to the date of the 
inquiry. (4) In St. John’s Wood there was no scarlatina 
whatever down to the date of the inquiry, although the 
dealer there got five-sixths of his milk from Hendon. 
Were there any conditions in the farm operations parallel 
to these special phenomena? And first, was there any 
new condition pertaining to the cows coincident with 
the milk producing scarlatina at the end of November in 
four districts, continuously in South Marylebone, and 
after a break in the other three, while this condition was 
absent in the case of the cows supplying the St. John’s 
Wood dealer? A tedious inquiry into such circumstances 
as the food, calving, health, arrival and departure of cows 
proved barren of result; nothing could be heard of for 
some time that was new or changed. But at last it 
appeared that on November 15 three newly-calved cows, 
NATURE 
purchased in Derbyshire, had com: on the farm, and four 
from Oxfordshire on December 4. The practice of the 
farm was to isolate or quarantine new arrivals for exami- 
nation for a week or ten days, and then to admit them 
into the stalls with the others. The cows on the farm at 
this period numbered 90 or 100, distributed in unequal 
numbers in three sheds, called the large, middle, and 
small sheds. The supply of the milk from the large. 
shed went to South Marylebone only; that from the 
middle shed partly to South Marylebone, partly to 
Hampstead and St. Pancras; and that of the small shed 
to the two latter places and to St. John’s Wood. So far 
we have this coincidence between the doings at the farm 
and the incidence of the disease—that the latter broke out 
after the time that the milk of the Derbyshire cows was 
added to the general stock, in three districts supplied 
from the farm ; and that St. John’s Wood, which did not 
receive any milk from the new arrivals, was free from 
scarlatina. 
We have now reached what may be called the third 
stage of the case. In the first, what Mr. Power calls a 
“ notable,’ and what lawyers perhaps would call a “ vio- 
lent,” presumption had been made out against the Hendon 
dairy ; in the second, a weaker presumption had been 
established against the Derbyshire cows which had been 
added on November 15, and whose milk began to be dis- 
tributed to the three affected districts, and not to St, 
John’s Wood, a few days later. But then, the facts of a 
continuous and increasing attack in South Marylebone, 
and the intermission of about ten days in St. Pancras 
and Hampstead, had to be accounted for, if the case was 
to be made out conclusively against the incriminated 
dairy. To deal with these, Mr. Power reversed the pro- 
cess hitherto pursued, which was that of pure induction 
from observed facts. He now employed the a prioré 
process, ana argued thus :—Taking the fact of uninter- 
rupted progress of the disease in South Marylebone, and 
of the lull of ten days in the other two, if the dairy at Hen- 
don be the cause of the outbreak, and if, as is most pro- 
bable, the different results produced by the milk from the 
same cows was due to a difference in the relation of the 
cows themselves within the business of the farm, then 
we should find at the latter—(1) a change in the manner 
of distributing the milk of the Derbyshire cows, and this 
probably consisting in placing them, or one of them, in 
the ‘‘large shed,” from which South Marylebone was 
supplied ; (2) about the second week in December (the 
date of the recrudescence of the disease in St. Pancras 
and Hampstead), some of the Derbyshire or of the Ox- 
fordshire cows, or some other cows which had been in 
close relation with them, were probably transferred to the 
“middle shed,” from which these two districts were, it 
will be remembered, supplied; (3) as St. John’s Wood, 
which was supplied from the “small shed,” was free from 
scarlatina, it should be found that none of the new cows, or 
any other cow in close relation with them, had been placed 
there. Now, were any arrangements at the farm found 
corresponding with any or all of these @ gréoré conclu- 
sions or probabilities? What was found on investigation 
was this: (1) The Derbyshire cows had _ been transferred 
towards the end of November into the “large shed ” (the 
source of the South Marylebone supply), and remained 
here at the date of the inquiry ; (2) the four Oxfordshire 
cows were transferred about December 11, two into the 
“Jarge shed,” and two into the “middle shed” (St. 
Pancras and Hampstead supply); (3) at no time had 
either the Derbyshire or Oxfordshire cows been trans- 
ferred to the “small shed” (St. John’s Wood). Here, 
then, both by positive and negative evidence, the presence 
of scarlatina in certain London districts was associated, 
first, with a particular dairy, and secondly, by a series of 
parallel events, with certain cows within that dairy. | 
Power, having reached this point, felt justified in assuming, 
until anything to the contrary should appear, the presence 
[August 26, I 886 
Mr, © 
ee Se 
