Bacteriology 102:
Salmonella – General Aspects and Nomenclature

John L's Bacteriology Pages >
Bact.102 Website–Fall 2006 >
Page Last

By John Lindquist.

Periodically updated! Based originally on material in the Laboratory Manual for the Food Microbiology Laboratory (1998 edition, edited by John L. and published at the University) which was used in Bacteriology/Food Science 324 at the University of Wisconsin – Madison.

This page is presently relevant to a Virtual Exercise on Salmonella in the UW-Madison Microbiology 102 course and is not intended to be authoritative for applied or research purposes.

The genus Salmonella is of great concern to the food industry. Processed foods are not permitted to contain any Salmonella cells. The reason for this "zero tolerance" is that salmonellae are responsible for severe, acute gastroenteritis. Salmonella is responsible for a major portion of the cases of gastroenteritis each year in the United States.

Salmonellae are widespread in nature. They are commonly found in the intestinal tract of mammals, birds and reptiles. Pork and beef are often found to harbor salmonellae. Poultry continues to be one of the primary reservoirs of the organism. Any raw poultry product should be assumed to contain salmonellae and therefore must be handled appropriately. Eggs are contaminated with salmonellae when laid. Pasteurization of the egg is designed to inactivate the organism, though one should not casually assume eggs are then necessarily free of it. In recent years, there have been numerous outbreaks of gastroenteritis credited to Salmonella-contaminated eggs.

Taxonomically, the species concept has been applied with difficulty to members of this genus over the years. Through much of the 1970's and 80's, the three-species concept was generally utilized with the recognition of S. typhi, S. cholerae-suis and S. enteritidis. The last-named species, whose name was derived from a very early-established serovar (serotype), was basically a dumping ground for the vast majority of strains isolated, generally from cases of gastroenteritis.

The name "Arizona" appears to have been applied in one form or another to a group of Salmonella-like organisms since the late 1930's according to the 7th edition of Bergey's Manual (1957) where it was called Paracolobactrum arizonae. Serological similarity with Salmonella was usually a noted feature along with the atypical (for Salmonella) tendencies to ferment lactose and liquefy gelatin. Later names included Arizona arizonae, Arizona hinshawii and Salmonella arizonae before being totally absorbed into the modern Salmonella scheme as two of the recognized subspecies.

Several genetically and biochemically-defined subspecies of Salmonella are recognized. However, the "handle" for any isolated strain of Salmonella is a further subdivision, the serovar – also known as serotype – a designation which reflects the testable antigenic makeup of the organism, involving the identification of the cell wall ("O") and flagellar ("H") antigens. Each identified strain belongs to one of over 2000 recognized serovars. All serovars are designated with an antigenic formula, a series of numbers and letters which refer to the recognized antigens. Most serovars also have names which have been often written for convenience as if they were (but they really are not!) species names – for example, Salmonella typhimurium. The present convention is to write the serovar name without italics, starting with a capital letter. So, a correct (but cumbersome) way to indicate the aforementioned serovar is Salmonella enterica subsp. enterica ser. Typhimurium. A more convenient, shortened version would follow this form: Salmonella Typhimurium.

Although the term serovar is used predominantly in this discussion, the admonition by Salmonella and nomenclatural authorities to use "var"-suffixed terms has been resisted by many bacteriologists who have preferred to stick with the more descriptive, familiar and euphonious term serotype.

The following review explains host-adaptation as it relates to Salmonella: Andreas J. Baumler, Renee M. Tsolis, Thomas A. Ficht, and L. Garry Adams. 1998. Evolution of Host Adaptation in Salmonella enterica. Infect. Immun. 66: 4579-4587.

The new, online List of Prokaryotic Names with Standing in Nomenclature includes considerable discussions of Salmonella taxonomy and nomenclature, citing judicial opinions brought about by various requests for official clarification of these matters. An authoritative summary on this page includes a proposal for a third Salmonella species, the name of which (S. subterranea) was validly published as such but turns out to be more closely related to one or more other enteric genera.

The table below lists the various subspecies within the recognized species of the genus Salmonella. The first subspecies embraces the vast majority of isolated serovars. Most of the serovars of the first subspecies are responsible for salmonellosis (gastroenteritis) in humans, a prime concern to the food microbiologist. These organisms are virtually identical genetically, and they tend to possess the same phenotypic characteristics. Exceptions to this rule occur with "host-adapted" serovars – i.e., those which have become adapted in their evolution to certain animals with (often) a consequent and significant shifting of certain physiological properties. They are sometimes referred to as bioserovars or bioserotypes, and it can be tempting to raise them to species level because of their pathological significance, a notable example being the serovar Typhi. (This is the feeling that keeps the pathogens named Bacillus anthracis and Yersinia pestis from being absorbed into other very closely-related and earlier-recognized species.) Following is a list of some host-adapted serovars; major phenotypic variations are highlighted:

  • Serovars Gallinarum and Pullorum cause fowl typhoid and pullorum disease, respectively, in poultry. These serovars are unlike typical salmonellae in being non-motile. They are two biochemically-distinct entities which possess the same antigenic formula, and use of the two separate names makes an exception to the formula/name rule. So, it is becoming the practice to place them together under one serovar designation and then differentiate them at a lower level into two biovars (biotypes) as discussed here.
  • Serovar Typhi causes typhoid fever in humans; this is the classic "enteric fever." Gas is not produced during fermentation.
  • Serovars Paratyphi-A and Paratyphi-B cause milder forms of enteric fever in humans. Most strains of the former do not produce hydrogen sulfide.
  • Serovars Cholerae-suis and Typhi-suis cause severe disease in hogs.

Salmonellosis results from the ingestion of viable Salmonella cells. Generally, about 108 cells must be ingested, although there are reports of as few as 100 cells being able to cause gastroenteritis. The disease is referred to as a toxico-infection, as one must ingest viable cells and have these cells colonize the small intestine. During their growth in the intestine, they release an enterotoxin, resulting in severe vomiting and diarrhea. Onset of symptoms occurs about 12 hours after ingestion of the contaminated food with symptoms lasting 2 to 3 days. The disease is generally self-limiting.

The isolation and identification of Salmonella can be a very time-consuming and expensive process and includes the use of a number of selective and differential media. Foods should not be released by the manufacturer until declared free of salmonellae. This requires holding the product, at consumer expense, for a protracted period of time. Research continues to develop rapid, reliable methods to detect salmonellae.

The method traditionally used by the Food and Drug Administration begins with a nonselective enrichment in lactose broth. A selective broth enrichment follows on the next day. On the third day, the selective enrichment is streaked onto several different selective-differential plating media to initiate the isolation of any salmonellae present. (The sample can be plated directly if a large, active population of salmonellae is suspected.) Should likely colonies be found, they are subcultured into Triple Sugar Iron (TSI) Agar, Lysine Iron Agar (LIA) and Brain Heart Infusion (BHI) Broth. If the results of the TSI Agar and LIA indicate the probability of Salmonella, the BHI Broth culture is employed in the serological confirmation of the presence of the organism with the use of polyvalent flagellar antisera. Tests for identity of cell wall antigens may also be done, using cells from the TSI Agar. However, certain other genera contain strains which share the same cell wall antigens as many salmonellae, so this test is not genus-specific. Further confirmation involves tests with various differential media; this is especially useful when a non-flagellated strain is encountered. The entire process usually takes about six days to complete.


The antigenic formula summarizes the unique combination of testable antigens associated with a serovar. As an example, the antigenic formula for the serovar Typhimurium – 1, 4, 5, 12 : i : 1, 2 – is constructed from the following sets of detected cell wall (O) and flagellar (H) antigens:

O antigens detected: 1, 4, 5 and 12
H antigens detected – phase 1: i
H antigens detected – phase 2: 1 and 2

By "phase 1" and "phase 2" we mean two sub-populations of cells where the cells in each possess a certain antigen (or set of antigens) associated with their flagella. In testing for the various H antigens possible in a culture of a Salmonella isolate, the antigens of the majority sub-population are initially detected by the use of individual, specific antibodies. By spot-inoculating the culture on a solid medium containing these antibodies, cells of the majority sub-population can be immobilized while cells of the minority sub-population (if present) can swarm out onto the medium (as their flagella are not hindered by the antibodies in the medium) and proliferate, such that they can be isolated and tested for their H antigens. This has been termed the "phase-reversal technique."

The example above shows a "diphasic" serovar. Examples of "triphasic" and "quadriphasic" serovars are included in the table below.


In the following procedure, the test for cell wall ("O") antigens can be done on an isolate likely to be Salmonella. With a wax pencil, circular areas are marked off on the surface of a glass slide. These marks should be drawn heavily in order for the circular areas to contain cell suspensions which must not be allowed to run into each other or off the slide. After a drop of cells suspended in saline is placed in each circle, a drop of antiserum is subsequently added to each. (Alternatively, the antiserum drops can be placed on the slides first, and the cells can be suspended directly into the antiserum.) Where there is a reaction between antibodies in the antiserum and their homologous antigens on the cell wall of the bacteria, the cells will clump together ("agglutinate"), and the drop will appear to contain many small particles. The reaction is best observed from underneath the slide, and this can be accomplished with a mirror or by placing the slide in a petri dish and then holding the dish above eye level. In this photo, one drop is seen to contain agglutinated cells, and the other retains its original milky appearance.

The test for flagellar ("H") antigens is done in tubes, and we have no photo of this test as yet to show here. Strains of some other genera (e.g., Citrobacter) may possess one or more of the same O antigens of Salmonella; the H test is much more specific for Salmonella identification.


The table below is based on the system wherein the primary species of Salmonella is called Salmonella enterica. (A competing system following the traditional rules of nomenclature designates the same species as Salmonella cholerae-suis and the first subspecies as Salmonella cholerae-suis subsp. cholerae-suis.) Over the past couple decades, Salmonella bongori has become the second bona-fide species, having been upgraded from subspecies status as can be seen in the ninth edition of Bergey's Manual of Determinative Bacteriology (1994).

Subspecies are defined and differentiated biochemically and genetically. Also termed "subgroups," the subspecies names in the table are preceded by the numerical subgroup designation.

Several host-adapted, biochemically aberrant serovars ("bioserovars") which do not have unique antigenic formulae (and must then be distinguished by their biochemical reactions) are shown in this table – i.e., Pullorum, Gallinarum, Paratyphi-C, Cholerae-suis and Typhi-suis.

Irrespective of subspecies designation, the serovars are grouped according to their O antigens: Those which share O antigen 2 are in Group A, those which share O antigen 4 are in Group B, those which share O antigens 6 and 7 are in Group C1, etc.

Lists of serovars can be found in various references such as Bergey's Manual of Systematic Bacteriology, Second Edition, Volume 2, Part B (2005) and the 1998 edition of the Difco Manual. For the ultimate list according to a controlling authority, the WHO (World Health Organization) Collaborating Centre for Reference and Research on Salmonella periodically publishes the Antigenic Formulas of the Salmonella Serovars.

subspeciesantigenic formulaserovar name
1.  subsp. enterica1,2,12:a:1,5Paratyphi-A
 6,7:y:e,n,z15:z47:z50Mikawasima 1
 13,23:d:1,7Grumpensis 2
 30:i:e,n,z15Mjordan 3
 47a,47b:z45:z4,z23:z6Bere 4
2.  subsp. salamae1,9,12:l,w:e,n,xDaressalaam 5
 11:b:1,7(none given)
3a.  subsp. arizonae 651:z4,z23:-(none given)
3b.  subsp. diarizonae 640:k:e,n,x,z15(none given)
4.  subsp. houtenae11:z4,z32:-(none given)
 43:z4, z23:-Houten 5
5.  subsp. bongori 748:z35:-Bongor 5
6.  subsp. indica1,6,14,25:a:e,n,xFerlac 5

1  An example of a quadriphasic serovar.
2  An example of a serovar named after the individual from which it was isolated first – in this case a person (hospital patient?) who was referred to as being grumpy (or perhaps as Grumpy). "Grumpensis" is Latin for "from Grumpy."
3  An example of a serovar named to honor a specific individual – in this case, basketball legend Michael Jordan. The name was seen in the press as Salmonella mjordan.
4  An example of a triphasic serovar.
5  Names for serovars other than those in the first subspecies have been deprecated, and only the antigenic formulas are recognized as serovar designations; the old names for these particular serovars are shown on this table.
6  Set apart in some older classifications as a separate genus (Arizona) or as a separate species of Salmonella (S. arizonae). Subgroup 3a serovars are monophasic; subgroup 3b serovars are diphasic.
7  The sole subspecies of Salmonella bongori.

Selected Groups of Bacteria
Bacteriology 102 Website
This page was last modified on 3/23/15 at 6:00 PM, CDT.
John Lindquist, Department of Bacteriology,
University of Wisconsin – Madison

An older edition of this web page which was prepared in the early
morning of Sept. 11, 2001 is archived here. While in preparation, the
major national news item concerned Michael Jordan's future in basketball.