The journal that is now Epidemiology & Infection was founded in 1901 as the Journal of Hygiene ‘to fulfil a definite purpose by serving as a focus to English-speaking investigators for works in physics, chemistry, physiology, pathology, bacteriology, parasitology and epidemiology in relation to hygiene and preventive medicine’. In 1987 the name of the journal was changed to Epidemiology & Infection. The reasons for this change were articulated by John Pattison [1] on one of the few occasions that the Editor has directly addressed the readership and formalized the historic shift towards microbiology and epidemiology.

For over two centuries yellow fever (YF) had terrorized those visiting and living close to endemic areas. Prompted by continued imports and its impact on US troops in the Spanish–American War of 1898, US Surgeon-General George Sternberg sent a small Commission to investigate the problem in US-occupied Cuba. It comprised James Carrol, Jesse Lazear and Aristide Agramonte; its leader was Surgeon-Major Walter Reed. Whilst in Cuba they collaborated with Cuban Juan Guitéras of Havana's Sanitary Department and Surgeon-Major William Gorgas, American chief of the island's Sanitary Department.

‘The opportunities of a pathologist at a large Eastern Hospital are many; but his time for research work is short, and his conveniences are few.’ So begins this landmark paper written by Alfred Whitmore and published in the journal in 1913 [1]. He goes on to demonstrate just how well that short time can be used by someone with the energy, intelligence and scientific rigour in order to make the most of those opportunities. I first read this paper 15 years ago and have re-read it many times since, usually reflecting on how remarkably little our knowledge of melioidosis has advanced since 1913. I notice something new each time I read it and, most importantly of all, each time it is an absolute pleasure to read. How many modern scientific papers leave one feeling the same way?

Towards the end of the 19th century there was a need for standardized tests to determine the efficiency of disinfectants against the newly discovered bacterial pathogens. Methods had been developed in which disinfectants were tested on organisms soaked into thread or coated onto garnets. However, more appropriate were tests based on contact between dilutions of disinfectant and suspensions of organisms, which could be subcultured to detect residual infectivity.

Diagnostic microbiology, first by the visualization of organisms followed by culture of pathogenic organisms, developed at the end of the 19th and the beginning of the 20th century. The detection of specific antibody as a means of diagnosing infection followed. One of the most enduring of these antibody assays has been the complement-fixation test, which is still in use 40 years after its development by Bradstreet and Taylor in 1962 [1]. Despite the development of more sensitive assays such as RIA and ELISA, the serological diagnosis of many respiratory infections is still often achieved by means of the complement-fixation test. Some of these pathogens such as influenza viruses are also diagnosed by culture although obtaining acute phase specimens for culture is problematic. Other organisms such as Mycoplasma pneumoniae and Coxiella burnetii, although cultivable are seldom diagnosed by this means because most diagnostic laboratories are not able to perform culture routinely. Diagnosis of these treatable infections is usually made by use of the complement-fixation test.

The nature and frequency of waterborne disease changed substantially during the course of the 20th century [1]. In the early decades of the century, the primary diagnosed cause of waterborne disease in both the United Kingdom and the United States was typhoid fever. It was well known at that time that the primary source of infection was faecal pollution of drinking water. It was against this background that methods for detecting faecal pollution were sought.

The routine title of the long ‘note’ by Glenny and Südmersen reproduced here [1] hides two of the most significant findings in theoretical and applied immunology: a brief description of diphtheria toxoid and a comprehensive account of the primary and secondary immune response. The introduction of antitoxin treatment of diphtheria was heralded, in 1896, without exaggeration as ‘the most important advance of the [19th] Century in the medical treatment of acute infective disease’ [2]. However, there were problems with the standardization and potency of antitoxin. The former was solved by Ehrlich who, in his analysis of toxin–antitoxin interaction, postulated the existence of ‘toxoid’, a non-toxic component of toxin which combined with antitoxin [3]. The latter was solved by the general observation that repeated injections of gradually increased doses of toxin induced increasingly potent antitoxins, an important basis upon which the present paper was founded.

Measles has been known as a highly contagious serious infection in man for centuries. After a prodromal fever, patients often experience coryza, cough and conjunctivitis followed by a blotchy rash. The disease is more severe in infants and adults with death resulting from encephalitis or pneumonia. Malnourished and immunosuppressed people are also more prone to serious complications and death. Subacute sclerosing panencephalitis (SSPE) is a rare complication, occurring as late sequela several years after measles infection at a rate of 1–5 per million cases.

During its first century the journal attracted well over 100 papers on poxviruses. During the last 50 years this was probably due in part to the fact that Allan Downie, the world's leading poxvirus authority helped to edit the journal from 1951 to 1982. This expertise and association was acknowledged by the publication of a special academic tribute of 12 papers, including 5 on poxviruses, in 1982 (89: 353–478).

(Reprinted from: Hayes W. The discovery of pneumococcal type transformation: an appreciation. J Hyg 1966; 64: 177–184.)

Griffith F. The significance of pneumococcal types. J Hyg 1928; 27: 113–159.

This paper [1] is a milestone in the literature on infectious diseases. To put it in context, we recall that the late 19th century saw the construction of the germ theory, and its ultimate acceptance by the medical profession. The massive research effort led by Pasteur and Koch and their followers demonstrated a variety of infectious agents, catalogued their properties, and traced their pathogenesis in infected hosts. An understanding of the behaviour of infections in populations came only later, in the early 20th century, exemplified in the work of Ross on malaria [2], which was extrapolated to all infections in his ‘theory of happenings’ [3], and of Hamer on measles [4]. But there remained a tension between those who viewed infections from the perspective of the laboratory, with its emphasis upon biological properties, and those who viewed disease from the perspective of population statistics [5], which lent itself to more abstract and mathematical descriptions of epidemiological patterns. Fierce battles were waged between these disciplines, as between Almroth Wright and Karl Pearson on the subject of typhoid vaccination [6, 7].

The Stonehouse meningitis survey took place in 1986 [1]. The context was a substantial and highly focal outbreak of meningococcal disease affecting the population of southern Gloucestershire and particularly the town of Stroud and its adjacent communities – a population of some 90000 individuals in the west of England. Almost all cases of meningococcal disease were due to a serogroup B, type 15, subtype P1.7,16 sulphonamide-resistant strain that had not been documented in Gloucestershire (and only rarely in the United Kingdom) prior to 1982, the first year of the outbreak. As is often the case at the start of an outbreak of meningococcal disease in a temperate country, the attack rate rose particularly in older children. For the population of Stroud as a whole the meningococcal disease attack rate rose from approximately 1·0 cases per 105 to approximately 8 per 105 population.

Predicting the impact of infectious diseases on the well-being of the community is a cornerstone of identifying effective prevention, control and support. One only has to reflect on the last few years in the United Kingdom to see the impact mathematical modelling has had on public and government, with controversy around the likely numbers of sufferers from new-variant Creutzfeld–Jakob disease, human immunodeficiency virus, and continuing debate as to whether to use vaccine to support control of foot-and-mouth disease. Over the last 20–25 years, Epidemiology and Infection, and before it the Journal of Hygiene, have published many of the sentinel papers in the mathematical modelling of infectious disease, both in humans and animals. The discipline has advanced from relatively simple analyses to the most complex assessments whose underlying mathematics and statistics almost certainly exceed the comprehension of all but a few microbiologists and public health specialists. The depth of the analysis does not obscure the key messages, however, for the epidemiology of infectious disease and its control, and a Special Article in the journal in 1988 overviewed its contribution [1].

Epidemiology & Infection probably attracts more papers on mathematical modelling of infectious diseases than does any other epidemiology journal. The most important modelling papers published in the journal were probably those of Anderson and May during the 1980s, which laid the foundations for much of the subsequent modelling work carried out by themselves and their colleagues. Since the start of their partnership, they authored 17 articles between them in the journal, including work quantifying the effect of different vaccination strategies against measles and rubella [1, 2], on the epidemiology of rubella in the United Kingdom [3], and on the effect of age-dependent contact between individuals on the critical level of vaccination coverage required for control [4]. The latter work, published in 1985, was particularly important, since it described methods for incorporating realistic assumptions about (heterogeneous) mixing between individuals into models, an issue which was beginning to be addressed in the mathematical literature but which had not yet reached many epidemiological journals. Other important modelling work published in Epidemiology and Infection includes that of McLean et al. (reproduced in this edition) on the control of measles in developing countries [5, 6], and by Garnett and Grenfell on the epidemiology of varicella zoster in developed countries [7, 8].