It has been shown that S. schaudinnii will easily pass through Berkefeld V filters under a negative pressure of considerably less than one atmosphere, but B. pyocyaneus was always found in the filtrates in association with the spiro-chaetes. Attempts to pass S. schaudinnii through Berkefeld N filters were uniformly negative.

Inoculation experiments on native volunteers made with (a) filtrates containing S. schaudinnii and B. pyocyaneus, (b) a mixture of these organisms combined with B.fusiformis, and (c) an emulsion of B.fusiformis in broth, did not give rise to ulceration.

1. Samples of milk from 183 cows were examined, and seventeen (9·2 per cent.) yielded Br. abortus.

2. Two hundred and two samples of butter, margarine, cheese, and icecream failed to yield Br. abortus, although samples of locally prepared butter and cheese both contained tubercle bacilli.

3. Examination of the serum from slaughtermen and men working in allied trades showed some evidence of latent infection, but no indication of much real illness due to Br. abortus.

4. Examination of the sera of sheep failed to show any evidence that these animals become infected with Br. abortus.

5. The examination of sera and organs from pigs failed to show any evidence that the porcine type of Br. abortus occurs.

It is now generally accepted that certain bacterial toxins are able to give rise to fatty degeneration in tissues and may bring about marked changes of this nature in the heart, liver, kidneys and other organs. When a cell is damaged a state of cloudy swelling or of parenchymatous degeneration generally occurs. The fat is distributed throughout the cell as fine globules and the condition is known as fatty degeneration. Numerous workers (see References) are agreed that many toxins, particularly diphtheria toxin, may cause a widespread damage to cells of the body in this way. The processes involved in the deposition of this fat have led to much discussion and speculation, but do not concern us here.

1. That the epidemic was due to infection of the water supply is borne out by the following points:

(a) The explosive character of the outbreak as shown in Graph II which is characterised by a rapid rise to peak at the end of third week followed by a sharp fall and a gradual tailing off to the end of the epidemic.

(b) The widespread distribution of cases (see Map), no area supplied by water from the city having escaped.

(c) The rapid decline which followed the adoption of appropriate measures to deal with the water supply.

2. 255 persons in the city area (including the 12 military cases) were definitely affected (equivalent to a rate of 3·1 per 1000 of the population). As shown by the rise in the death-rate for the rural area it is apparent that a far greater number of persons were actually attacked.

3. The outbreak had its origin in a defective manhole in the Mental Hospital grounds whereby the sewage overflowed into a rubble drain and infected the water supply. This defect had existed for twenty-five years, and an epidemic in the hospital fifteen years previously established the reservoir from which infection was disseminated through the city.

4. The spread of typhoid was favoured by unsatisfactory arrangements for the filtration of the water supply. This is borne out by the endemic character of the disease prior to the installation of an effective filtration plant.

5. The institution of this plant was followed by an immediate reduction in the incidence of the disease until, as at the present time, water-borne typhoid has been practically eliminated.

6. The value of such a plant combined with chlorination and bacteriological examination has been shown.

1. The commonly occurring bacteria may be divided into four groups on the basis of their behaviour at various temperatures. These are (i) the Staphylococci, not growing below 10° C, (ii) most strains of B. coli, B. proteus and Micrococci ceasing growth in the range 5−0° C, (iii) some strains of B. proteus etc. capable of growth at 0° C, (iv) many strains of Achromobacter, Pseudo monas, and various yeasts growing rapidly at 0° C. (sometimes in 5 days), and down to about −5° C. on unfrozen media.

2. Bacterial growth on frozen media has not been observed below −3° C.

The author is indebted to Dr G. S. Graham-Smith, F.R.S., who read the manuscript, to Mr W. A. Bmpey for the supply of cultures, and to Mr J. Barlow for assistance in the experimental work.