The 1914 U.S Treasury Department standards specified that an acceptable drinking water must not show presence of B.Coli in more than one of the five 10-ml portions of the sample tested. The terminology B.Coli was retained in the 1925 standard, but the examination of a series of samples rather than appraisals of individual samples was stressed. This realization of the statistical nature of a set of observation was a significant advance. By 1942, the term had changed from B.Coli to “coliform group.”

The membrane filter (MF) procedure was first included as an alternate to the multiple-tube procedure in the 1962 standards.Standard Methods for the Examination of Water and Wastewater defines the coliform group as including “all of the aerobic and facultative anaerobic, Gram-negative, non-spore-forming rod-shaped bacteria which ferment lactose with gas formation within 48 hours at 35 ° C.”The coliform group includes both fecal and non-fecal bacteria. They are the predominant organisms that are found in the intestinal tract of man and other animals, both warm-blooded and cold-blooded. Certain non-fecal soil bacteria, such as Aerobacter aerogenes, are also included in this group. Although bacteria procedures have been developed which measure only the fecal coliform, health-oriented personal agree that the examination of drinking water should be based on all bacteria of the coliform group. Presence in drinking water of significant numbers of any members of the coliform group indicates either deficiencies in treatment of the water or inadequate protection of the source of untreated water. Laboratory procedures for both the MPN dilution and the MF methods for coliform bacterial examination of water are given in Standard Methods.

The MF procedure has the advantage of securing the results in less time. It also requires less work and less incubator space. Geldreich et al., however, note that both skillful and careful laboratory work is required to secure satisfactory results.

A heath benefit frequently overlooked when the quality of water is improved is sometimes known as “Haze's theorem.” In 1904 Allen Hazen presented a quantitative expression for the death rates from other diseases as related to those for typhoid fever. As it was expressed at the time, when one death from typhoid fever is avoided by the use of better water, two or three deaths from other causes are also avoided. This proposition also was noted by other and is sometimes referred to as the “Mills-Reincke Phenomenon,” named for Hirm F.Mills, a civil engineer of Lawrence , Massachusetts , and Dr.J.J.Reincke of Hamburg, Germany. These gentlemen reported somewhat similar information at virtually the same time. Two factors that may be presumed to have contributed to this phenomenon are, : first, when the human system is freed from having to resist typhoid infection, it has greater resistance to other diseases; and second, better treatment results in water relatively free from organisms causing diseases other than typhoid. At Hamburg,Germany , for example, for each decrease in death from typhoid fever after installation of filtration, there were 15.8 fewer deaths from other causes. At Lawrence , Massachusetts , this ratio was 1:4:4. At Lowell , Massachusetts , it was 1:6.0. At Albany , New York , it was

1:5.5. Statistics showing the decrease in the death rate from typhoid fever accompanying the advent of water purification practice usually fail to emphasize this additional benefit.