the period of advance. Two patients only had secondary rises of temperature. In distribution the rash was in every detail typical of smallpox as also was the order of its appearances. The elements of the rash appeared to pass through the various stages rather quickly, and were generally much more superficial and much smaller than those usually met with in smallpox. Very often also the lesions aborted prematurely. While the usual shape was regularly circular, some patients showed rashes the elements of which were very irregular in outline. The rash was scanty in some patients, and very profuse in others. A few patients showed some confluence of the lesions on the face, but no case could properly be described as one of confluent smallpox. There would appear to be no doubt that the type of smallpox was exactly the same as that which has been so common in England during recent years. No evidence was obtained making it possible to estimate the length of the incubation period. In determining the probable date of infection, the incubation period was always assumed to be twelve days." In view of the mild character of the disease it was deemed desirable to confirm by vaccination the diagnosis of all cases admitted to hospital. Such cases as presented special difficulty of diagnosis or were of particular importance from the standpoint of epidemiology or administration were also investigated by means of the Gordon vaccinia agglutination test.

6. The Gordon Test.--The Board have been furnished with the following notes by Professor Tulloch on the use of the Gordon test during the outbreak. The medical officer of health states that he found the test quite reliable and very helpful.

Owing to the mild nature of smallpox which is occurring at present in different parts of the country, difficulty in diagnosis is bound to arise. Any test which could definitely establish a differential diagnosis between variola and varicella would be of special value. In the early spring of 1927 a few cases of severe chickenpox occurred in Dundee, and, although on clinical grounds there was no doubt as to the nature of these cases, crusts were collected from them and the crusts examined by the vaccinia agglutination test of Gordon. An unequivocally negative result was obtained. Several weeks later there occurred in Dundee an outbreak of smallpox of a mild nature, and the first cases observed were investigated by this same test and gave equally unequivocally positive results. It was the intention of the medical officer of health to have as many cases of variola in the outbreak examined by this test as was possible, but an immediate supply of vaccinia agglutinating serum being obtainable only through the kindness of LieutenantColonel Gordon, C.M.G., C.B.E., F.R.S., Consultant Bacteriologist to St. Bartholomew's Hospital, London, and as this source of supply was strictly limited, it was decided on

consultation with Dr. Burgess to employ the test in the investigation only of such cases as presented difficulty of diagnosis, or were of special importance from the standpoint of epidemiology or administration. This limited application of the test was rendered necessary by the fact that the preparation of the test serum is a somewhat lengthy procedure. The type of cases examined could be categorised thus :

66

66

(a) Cases of chickenpox (which occurred during the period of the smallpox outbreak) exhibiting any atypical features.

66

(b) Cases of smallpox which were atypical, and notably those cases which were exceptionally mild. (c) The cases of smallpox in which contact with previous declared cases was remote or difficult to establish.

"(d) Cases of smallpox occurring in parts of the city remote from the area where the epidemic was centred.

(e) Any suspicious cases of chickenpox occurring in burghs or landward areas around Dundee.

In the investigation of all these the test proved valuable. Corroboration of the clinical diagnosis in the initial cases, assistance in the diagnosis of doubtful cases, and its aid in tracing remote contact in those cases occurring in areas of the city distant from the epidemic centre, considerably strengthened the position of the local administrative authorities in dealing with the epidemic.

"The rationale of the vaccinia agglutination test is based on the hypothesis-a hypothesis that seems from the laboratory investigation of Gordon to be valid that the virus of vaccinia is essentially the same as that of variola, and differs from it only in that the vaccinia virus has developed a peculiar degree of pathogenicity for bovines, while the variola virus has developed a like pathogenicity for the genus homo.

"If the view that variola and vaccinia really constitute a unity were the only instance of such relationship between viruses differing in certain characters, such view would be open to serious criticisms, but the relationship between vaccinia and variola is, according to recent work, of the same order as that between the bovine and human types of bacillus tuberculosis, or between the bacillus abortus of Bang and the microbacillus Melitensis.

"Vaccinia agglutinating serum is prepared thus:

A rabbit is vaccinated intradermally, multiple inoculations being made, with vaccine lymph, and when the reaction is at its maximum (about seven days) the animal is killed, the blebs scraped off and the material so obtained is triturated in physiological salt solution. This material so treated consists of broken rabbit epithelial cells containing

vaccinia virus, and when it is inoculated sub cutem, intraperitoneally, or intravenously into a rabbit, the animal does not react to the epithelial material contained in the extract because this has been derived from its own species, but it does react and produce anti-bodies to the vaccine virus contained in the epithelium. After a prolonged course of immunisation, the serum of such rabbits when mixed with the material against which it has been immunised gives rise to a precipitate, and such serum also precipitates vaccine lymph, extracts of crusts from cases of virulent smallpox, and extracts of crusts from cases of alastrim, but it does not precipitate extracts of crusts from cases of varicella or herpes zoster.

66

In conducting the tests, a suitable range of dilutions of the anti-vaccinia serum and a corresponding control series of dilutions of normal rabbit serum are distributed into tubes, and to each of these tubes is added a known quantity of saline extract of the crusts to be examined. Known positive controls of vaccinia extracts-rabbit or bovineknown negative controls of varicella extract are put up at the same time to ensure that the reagents are active, and are specific in their reactions. The mixtures of anti-serum, normal serum and extracts are incubated for 24 hours at 55° C. and readings are made with the agglutinoscope.

"There is seldom any difficulty in interpreting the results obtained, provided that adequate controls have been included in the test, and it may be stated that in our experience at Dundee the validity of the findings has in every instance been borne out by clinical and epidemiological enquiry.

"As to the utility of the Gordon test, it may be said that in the diagnosis of clinically typical and severe smallpox it is of but little value. On the other hand, in the investigation of mild and atypical smallpox (alastrim) and in cases of severe and atypical chickenpox, its value is very great indeed. The technique of the test is not unduly laborious, the results obtained are clear-cut, and the only difficulty which is encountered is that the necessary agglutinating serum is not easy to prepare."

7. Source of Infection.-The actual source of infection was not found. The type of smallpox was undoubtedly similar to that which has persisted in England during the last few years, and definite lines of communication between infected houses in Dundee and Durham County and South Wales, in both of which districts smallpox was prevailing at the time when the Dundee outbreak began, were found to exist. The evidence available was, however, not sufficiently complete to warrant a definite conclusion.

8. Conclusion. The continued prevalence of mild smallpox in England made it almost inevitable that sooner or later an out

break would occur in Scotland, and health authorities have realised the necessity of being on the look-out for the first signs. The mildness of its type gave the disease a considerable start in Dundee before it came to the notice of the medical officer of health. Once the existence of smallpox was known, however, the medical officer of health and his staff and the chief school medical officer and his staff lost no time in putting the necessary measures into force, with results for which they must be given the highest credit. Nor should it be overlooked that the speedy stamping-out of the disease could not have been effected by the officials without the co-operation and support of the local authorities. It was not until a month or two had passed without the emergence of any fresh case that the full effect of the measures taken was known, but it is now safe to say that these measures were entirely successful, while the amount of vaccination and revaccination carried out in the city may be expected to prove the best possible protection against the occurrence of another outbreak there. It is estimated that the cost to Dundee of dealing with the outbreak was in the neighbourhood of £5,800; and not only the citizens of Dundee but the whole population of Scotland have reason to be grateful that those responsible for the health of the city did not hesitate to take the measures and incur the expenditure necessary to free Dundee from smallpox. Should the occasion unfortunately arise, other local authorities could not do better than follow the example set in Dundee.

CHILDREN'S DISEASES.

9. Measles. The reduction in the number of cases of measles reported from the sixteen principal towns and the counties of Lanark and Fife was from 26.160 in 1926 to 16,692 in 1927; this conforms to expectation, as winters with heavy incidence of measles usually alternate with winters with comparatively light incidence, and the second half of the winter has more cases than the first half. In the winter of 1927-28 measles was very prevalent, 9,190 cases being reported in the last quarter of 1927 and 15,320 in the first quarter of 1928. Dundee and Kirkcaldy differed from the majority of places from which intimation of measles cases was received in that both had fewer cases than in 1926. The number of deaths in the sixteen principal towns fell from 657 in 1926 to 550 in 1927, and the death-rate per 100,000 persons living from 27 to 23. The corresponding figures for the whole country are 928 deaths (19 per 100,000) in 1926 and 660 deaths (13 per 100,000) in 1927, an indication of the greater severity of measles in the more densely-populated areas.

10. Whooping-Cough.-The benefit of a fall in the number of cases of measles was lost to some extent by a rise in the number of cases of whooping-cough from 5,526 in 1926 to 13,129 in 1927, 8,798 of which occurred in the first six months of the year. With the exceptions of Paisley and Kirkcaldy, the places from which

figures have been received all reported increases. The number of deaths in the sixteen principal towns rose from 292 (12 per 100,000) in 1926 to almost double that figure, 576 (24 per 100,000), in 1927. For the whole country the rise in deaths was less sharp, from 506 (10 per 100,000) to 850 (17 per 100,000).

11. Diphtheria.-There were reported 7,548 cases of diphtheria in 1927, an increase of 1,527 over the 6,021 cases which occurred in 1926. The heaviest incidence was in the fourth quarter, but each quarter had more cases than the corresponding quarter of the previous year. The increase was not confined to a particular area. The only towns showing decreases were Edinburgh, Aberdeen, Greenock and Hamilton, but the type of diphtheria prevalent seems to have been mild, for in spite of the larger number of cases there were two deaths less in the sixteen towns than in 1926, the death-rate remaining at 13 per 100,000, while for the whole country the number of deaths was five less than in 1926, the death-rate being 10 per 100,000.

12. Scarlet Fever.-In all four quarters of the year the incidence of scarlet fever was lighter in 1927 than in 1926, the figures for the two years being 10,609 and 13,667 respectively. To this fall of 3,058 all the reporting areas except Clydebank, Kilmarnock and Falkirk contributed. The number of deaths in the sixteen towns was 91, or less than half the number, 198, which occurred in 1926, the death-rates per 100,000 being four and eight in the corresponding years. The fall in the number of deaths is less marked for the whole country, the 1926 figure being 343-seven per 100,000-and the 1927 figure 200-four per 100,000.

The practice of testing persons for susceptibility to scarlet fever and to diphtheria and of immunising against the diseases those found susceptible continues to spread. The use of the Schick test for susceptibility to diphtheria and the active immunisation of susceptible persons will no doubt have been encouraged by the publication last year by the Medical Research Council of Dr. J. Graham Forbes's report on the Prevention of Diphtheria, in which attention is specially drawn to the extent to which these methods of prophylaxis have been applied in Scotland. The following statement by Dr. Rae, the county medical officer of Aberdeen, suggests how the immunisation of a rural population may be undertaken, and at the same time emphasises the need for care in the use of the immunising material

"The general practitioners ought, in my opinion, to be supplied with diphtheria and scarlet fever prophylactics free for use in children aged from one to six years, on the understanding that records would be kept of the numbers immunised. The only difficulty is that the material, when used once, deteriorates quickly, so that the practitioners who wished to use this method of prevention would require to collect their cases and inject (say) twenty at one time."