14 HISTORICAL OUTLINE 



inoculated. The development of our knowledge of these phenomena will be 

 discussed in Chapter 7. 



The demonstration by Loeffler and Frosch, in 1898, that foot-and-mouth disease 

 was caused by a virus which could pass through a porcelain filter, and was below 

 the limits of direct microscopical observation, opened a new field for investigation. 

 We now know many diseases which can be transmitted by filtered suspensions 

 of material obtained from infected animals, and in which therefore a filter-passing 

 virus is presumably concerned. 



There is nothing surprising in the fact that the years from 1900 to 1920, or 

 thereabouts, were for bacteriology a period of slower development as compared 

 with the riotous growth of the eighteen-eighties. The ground won had to be 

 consolidated. The previous advance had been a somewhat hasty affair ; and 

 many secondary problems had been left for more leisurely solution. Many bacteria 

 were very incompletely described. Many had been described independently by 

 different investigators, so that the same bacterium was masquerading under several 

 different names. Many of the earlier descriptions, especially of the anaerobic 

 bacteria, had been based on impure cultures. Little notice had been taken of 

 resemblances between bacteria, isolated at different times from different sources, 

 unless the practical application of the knowledge gained brought such resemblances 

 forcibly to the attention of some observer. Little was known about the distribu- 

 tion in nature of bacteria other than those concerned in the causation of disease, 

 or in some commercial or agricultural process, and even here the data were very 

 scanty. Bacterial ecology is, indeed, still almost an unexplored territory. In all 

 these directions the past thirty years have seen a considerable advance. Bacteria 

 have been studied more systematically. Fuller, and more accurate, descriptions 

 have been recorded and errors have been corrected. The labelling of the stock 

 strains of bacteria, scattered throughout the laboratories of the world, has been 

 more closely scrutinized. Many synonyms have been suppressed, and species 

 that had received several names have been shorn of all but one. The formation 

 of such collections as the National Collection of Type Cultures in this country 

 has provided a much-needed standard of reference. 



This re-survey of the bacteriological field on its qualitative side has been 

 associated with a great advance in our quantitative methods. The introduction 

 of bacteriological methods of analysis in the control of water-supplies, milk, and 

 so on, demanded standardized tests yielding numerical answers. Those first 

 employed were in most cases very faulty. There was a failure to realize many 

 of the technical sources of error ; and there was a still more general failure to 

 take into account the statistical principles involved in any sampling of this kind. 

 It is, indeed, only within recent years that a satisfactory liaison has been estab- 

 lished between the bacteriologist and the statistician ; and, even now, it is not 

 as general as it should be. It is not merely a question of the kind of analytical 

 test referred to above. The vast literature of immunity contains record after 

 record which is rendered meaningless by a neglect of the sampling errors involved 

 in working with small groups of animals. Reference to Chapter 43 will pro- 

 vide examples of the ways in which these errors may be avoided. 



In the last twenty years or so, there have been unmistakable signs that bacteri- 

 ology is on the march again. As always in experimental science it has been a matter 

 of technique. In this case the acceleration has followed the application to bacterio- 

 logy of the more exact methods of analysis developed by the chemist and the 



