History of the Calorie in Nutrition

Abstract

The calorie was not a unit of heat in the original metric system. Some histories state that a defined Calorie (modern kcal) originated with Favre and Silbermann in 1852 or Mayer in 1848. However, Nicholas Clément introduced Calories in lectures on heat engines that were given in Paris between 1819 and 1824. The Calorie was already defined in Bescherelle's 1845 Dictionnaire National. In 1863, the word entered the English language through translation of Ganot's popular French physics text, which defined a Calorie as the heat needed to raise the temperature of 1 kg of water from 0 to 1°C. Berthelot distinguished between g- and kg-calories by 1879, and Raymond used the kcal in a discussion of human energy needs in an 1894 medical physiology text. The capitalized Calorie as used to indicate 1 kcal on U.S. food labels derives from Atwater's 1887 article on food energy in Century magazine and Farmers' Bulletin 23 in 1894. Formal recognition began in 1896 when the g-calorie was defined as a secondary unit of energy in the cm-g-s measurement system. The thermal calorie was not fully defined until the 20th century, by which time the nutritional Calorie was embedded in U.S. popular culture and nutritional policy.

Introduction

The thermal units of g-calorie (4.184 J) and kg-calorie (kcal or capitalized Calorie; 4.184 kJ; Calorie is capitalized when the original text refers to a kg-calorie; the lower case word denotes a g-calorie) are so familiar in nutrition that one tends not to ask when they were first defined or how they entered common usage. The hypothesis of this paper is that the French word, calorie, had been coined and defined by 1824 and was originally used in the sense of a 0–1°C kcal (1–4). The Dictionnaire de l'Académie Française lists “CALORIE n. f. XIX^e siècle.” If so, this contradicts the statement that “Lavoisier served on the committee that developed the cm-g-s measurement system (CGS) which defined the “calorie” used today” (5). There is no documentation that the original metric commission defined a unit of heat. If Lavoisier's work is too early, other attributions of 1848–53 (3,6,7) are clearly too late, because the Calorie had already been defined in the 1845 edition of Bescherelle's dictionary of the French language (8).

Prior histories of the Calorie in nutrition do not discuss the origin of the calorie or kcal as units of heat (9,10). This article will show that use of the Calorie as a defined unit of heat developed concurrently with the metric system but not as a recognized metric unit and dates to work in chemistry or engineering no later than 1819–1824. A timeline of some key events is shown in Figure 1.

Figure 1

Timeline of 19th century events in the history of the metric system and calorie. BAAS, British Association for Advancement of Science.

In addition to its technical usage by scientists, the word Calorie entered the popular vocabulary across Europe and the United States by the late 19th century. For example, the Oxford English dictionary cites E. Atkinson's 1863 translation of Adolphe Ganot's French physics text (11) as the first occurrence of the Calorie in English. However, this physics text is not the source for the Calorie of nutrition. This article will outline the early usage and spread of the calorie as a heat unit from France to other countries during the 19th century.

Origin and usage of the calorie in France

The word calorie as a unit of heat seems to have been coined sometime between 1787 (1) and 1824 (12). Lavoisier studied specific heats of water and other materials and conducted some of the earliest experiments involving direct and indirect calorimetry (13). He named the calorimeter (calorimètre) by 1789 (14). Although Lavoisier was credited with coining “oxygen” and many new chemical terms, he did not include the calorie on his list of new words. Lavoisier's papers refer to calorique (caloric) and chaleur (heat) but not to the calorie as a thermal unit. At that time, “caloric” was regarded as a substance rather than a unit of heat. Lavoisier served on the 1791 Commission on Weights and Measures of the French Academy of Science and helped define the kg (5,15). A strict definition of the calorie would require metric units and Lavoisier used a quantity called the livre (about a pound) rather than the kg (16). He was executed by French revolutionaries in 1794 before the metric system was officially adopted in 1799 (13).

The original metric system of 1795–9 defined the base units for length, area, volume, capacity, weight (not mass), and money along with various prefixes (13). It was intended as a means of simplifying trade and did not include derived scientific units for energy, electricity, or magnetism. These secondary units were not defined until after the Metric Convention of 1875, when the Bureau International des Poids et Mesures (BIPM) was formed.

Several histories attribute the first usage of the calorie to sources that are too late. For example, Taton (7) indicates that Favre and Silbermann coined the term in 1852 (17). Others (6,18) also credit Favre for defining the calorie. However, the original publication states that the calorie was a well-known unit of physics (digital copies of L.N. Bescherelle's Dictionnaire National, Adolphe Ganot's Traite Elementaire de Physiques, and the original text from Favre and Silbermann's article on thermochemistry are available at the Gallica internet site: http://gallica.bnf.fr/ark:/12148/bpt6k34775c.table.):

Nous répétons que l'unité que nous avons adoptée est celle adoptée par tous les physiciens (physicists), c'est-à-dire la quantité de chaleur nécessaire pour élever 1 gramme d'eau de 1 degré, et que l'on appelle unité de chaleur ou calorie.

It is hard to assign priority to workers who state, “We repeat that the unit that we adopted is that adopted by all the physicists.” This is confirmed in the 1855 version (4th edition) of Ganot's basic physics text. This text contains a modern definition (11):

…c'est pourquoi on est convenu de prendre pour unité de chaleur, ou calorie, la quantité de chaleur nécessaire pour élever de zéro à 1 degré la temperature d'un kilogramme d'eau.

Because Ganot defines the calorie (not capitalized) as a 0–1°C kcal in relation to the heating of water without providing a reference, it seems evident that the calorie was well known. A French etymological dictionary lists the first occurrence of calorie as the 1842–3 volumes of Bescherelle's Dictionnaire national (19,20). The listing in the 1845 edition defines the calorie as Phys. Quantité de chaleur nécessaire pour élever un kilogr. d'eau un degré du thermomètre centigrade. The definition is similar to Ganot's more precise usage. Furthermore, the Dictionnaire historique de la langue française states that the word calorie was coined about 1819–24 and usage had become widespread by 1845 (12). Thus, French sources do not indicate that Lavoisier defined the calorie.

By 1824, the word calorie was being used as a unit of heat (1,3,4,12). In 1819, Nicholas Clément began giving lectures on the theory of heat and steam engines at the Conservatoire des Arts et Metiers in Paris (21). Clément was regarded as an industrial chemist or chemical engineer. Among the students was Sadi Carnot (3) and class notes are available that were taken by L.B. Francoeur and J.M. Baudot (2). Both Clément and Carnot subscribed to the caloric hypothesis, which stated that heat behaved as a material substance and that its total amount was always conserved (2). The notes show that Clément defined a large (grande) and small (petite) calorie by 1823–4. A definition is written in Baudot's notes of 23 December 1824:

La petite Calorie est la quantité de chaleur qu'il faut pour élever d'un degré la température d'un K.^me d'eau.

Assuming that the mass abbreviation refers to a kg, this defines a modern kilocalorie. Clément sometimes referred to a “grande calorie” as the heat needed to melt a kg of ice, which is ∼334 kJ/kg. This is different usage from the modern kcal, but the passage indicates that the calorie was known to engineers by this time (12).

Medard (4) suggests that Clément may have coined the word calorie around 1820 but agrees that this is not certain, because Clément did not publish the definition. Other than the hand-written course notes, the first published use of the calorie was probably in 1825 in an anonymous description of Clément's course in a local journal called Producteur (4). Carnot used Clément's definition of heat units but not the name calorie in Reflexions On The Motive Power of Fire (3,22). Other engineers began using the calorie by 1829 (4), but the word apparently did not enter physics texts until much later.

It is true that some scientists used heat units that would now be called g-calories in the 1850s. However, it was not until 1877–9 that Marcellin Berthelot stated that the “large” calorie equalled 1000 “small” 0–1°C g-calories and distinguished between them by capitalizing the abbreviation for the large Calorie (4,23). Although Medard states that the kcal was not introduced until 1935 (4), it was used in the context of daily energy expenditure in a U.S. medical physiology text from 1894 (24). A 14.5°C kcal was defined in German law in 1924 (3). It is not certain who first named the kcal.

The metric system In France

From 1795–9, a committee of the Académie des Sciences sought to define the meter as a unit of length related to an arc of the earth's circumference. The calorie could not be strictly defined without knowing the mass of water contained in a liter, which was based on a 10-cm cube. It is germane that the law of 1795 defined the g as “the absolute weight of a volume of pure water contained in a cube one-hundredth of a meter on a side at the temperature of melting ice” (25). This is important because the g-calorie eventually became a practical unit of heat, and the specific heat and mass of a volume of water depend on its temperature (26).

Because the kg, g, m, and cm were all defined from the start, it is ambiguous whether the original metric system should be called m-kg-s or cm-g-s. However, the kg was considered to be the base unit of weight in the 1795 metric system. This is because the charge to the commission was to define standards for weights and measures used in commerce, and scientific concerns were secondary. Note that the original definition specified a weight rather than a mass, meaning that the original kg was defined as kg-force, similarly to a pound. Not until 1904 was the newton defined as a unit of force so that the kg became a mass unit.

Because the g was not considered to be a base unit in the metric system (27) when the Calorie was defined, there was no need to add the “kilo-” prefix. The definition of a Calorie as the heat needed to raise the temperature of 1 kg of water by 1°C was the original usage and the kcal as 10³ g-calories was not introduced until sometime between 1877 and 1894 (23,24). Indeed, the thermal g-calorie was not defined as 4.184 joules until 1902 and it was a 17°C (not 15°C) unit (28).

The original metric system was short-lived as a French national standard. In 1812, Napoleon Bonaparte issued a decree to establish a “usual system” that changed metric values to conform to earlier weights and measures. Not until 1840 was a law passed to reinstate a metric system based on scientific standards (25). During this same period, European interest in thermodynamics, fuels, and electricity was high. Indeed, one of the reasons that Joule began his experiments on heat was a desire to reduce the cost of running equipment in the family brewery (29).

Metric advocates worldwide recognized the simplicity of basing units of measure on the decimal system, and the metric system was promoted at world expositions in London (1851) and Paris (1867). Also in 1867, the International Geodetic Association formed a committee to investigate the use of the metric system in scientific measurements. The system was made legal for commerce in Britain (1864) and the United States (1866) and became compulsory in Germany (1868) (15). The International Metric Convention of 1875 gave scientists impetus to define accurate standards in electricity, magnetism, and thermodynamics. This led to the introduction of the CGS in 1896 in which the erg, dyne, and joule were defined. In 1918, the newton was added as a unit of force to the m-kg-s system and the joule was defined as 1 N-m of work. The ability to define the joule unambiguously in base units (amp-s) led to its adoption as the SI (Système International) unit of energy (30).

During the 1930s, the BIPM convened the Consultative Committee on Thermometry to clarify standards of heat and W.H. Keesom served as president. In addition to reviewing the history of the calorie, Keesom summarized a proposal that the calorie should equal 1/860 watt-hours or 3600/860 joules (4.186 J) (3). From then on, any secondary thermal unit was to be defined relative to the joule rather than to the heating of water at any temperature. The 1948 General Conference also recommended discarding the calorie, because it cannot be derived directly from basic units. In 1954, the SI base units were adopted, and in 1970, the Committee on Nomenclature of the American Institute of Nutrition advised that the kilocalorie should be replaced by the kilojoule (kJ) in scientific publications (30,31).

The Calorie in German Physiology

The Calorie probably entered U.S. English because W.O. Atwater learned the term during studies in Germany, and not because it was defined in a newly translated physics text. Justus Liebig did not mention the calorie as such in his 1842 book on animal chemistry (32) or his paper on energy production from foodstuffs (33). However, he published J.R. Mayer's first scientific paper (34), which defined a mechanical equivalent of heat. Mayer self-published two intriguing papers that dealt partly with the efficiency of energy metabolism, which he estimated to be 15–20% (35,36). It was in the context of relating physical work against gravity (Fallkraft or potential energy; Fig. 2) to the energy supplied by foods that Mayer defined a kg-cal in 1846–8 (3,4,36). As translated by Lindsay (37) the quotation is:

When substances endowed with considerable chemical affinity for each other combine chemically, much heat is developed during the process. We shall estimate the quantity of heat thus set free by the number of kilogrammes of water which it would heat 1°C. The quantity of heat necessary to raise 1 kilogramme of water 1 degree is called a unit of heat, Calorie.

Figure 2

Caloric equivalent of work done against gravity. A mass of 427 kg falling 1 m against a 1 g gravitational field would increase the temperature of 1 kg of water by 1°C [redrawn from (57) with permission of the publisher].

The German text reads in part, “die Wärmemenge… nennt man Wärmeeinheit, Calorie.” Note that Mayer's definition, which occurred 48 y before the joule was introduced as a unit of energy, is essentially identical to the Calorie that is still used on U.S. food labels.

Mayer's views became known due to his dispute with Joule for priority in the discovery of the caloric equivalent of physical work (37), for which he was awarded the Copley medal in 1871, 1 y after Joule. Presently, it seems that an amateur has claim to the first definition of the Calorie in German science! Mayer may have learned of the Calorie because he studied in Paris and purchased a library that included French and German physics and chemistry texts of the period (38).

In 1860, Carl Voit visited Edward Frankland in England and brought a Thomson calorimeter back to his laboratory in Munich. The British Thermal Unit employed with the Thomson calorimeter was defined as the heat needed to raise the temperature of 1 pound of water by 1°F. Despite this, Voit began using the g-calorie in lectures on human calorimetry in 1866 and stated that daily metabolism of 1 male subject was 2.25 to 2.4 × 10⁶ g-calories, depending on prior diet (39). At that time, Pettenkofer and Voit (40) were using indirect calorimetry to study human and animal respiration and Voit was conducting nutritional balance studies (41). It was this juxtaposition between measuring human energy output and the energy content of foods by bomb calorimetry that gave the German school impetus to define daily energy intake and output in terms of calories.

By 1883–5, Max Rubner had published papers using the g-calorie to define heats of combustion for food and heat produced in respiration studies (42–44). In the same period, Henneberg and Stohmann were developing methods for the proximate analysis of livestock feeds at the Weende Experiment Station (45). Stohmann initiated calorimetric work in 1880 and conferred with Rubner concerning experiments in which the energy content of sugars, fats, and egg white was measured in g-calories (46). Note that different German scientists varied in using either the g-calorie or the Calorie.

The German school of physiology introduced proximate analysis and calorimetry to determine the least expensive sources of nutritionally balanced foods and animal feeds (39). This approach appealed to the American Wilbur O. Atwater, who visited Voit's laboratory in 1882–3 and conducted postdoctoral studies along with Rubner (47). In their 1896 report on nutritive values of foods, Atwater and Woods (48) specifically credit the Weende method of proximate analysis including fuel analysis as developed by Henneberg and Stohmann (45).

Atwater returned to the US and gave a presentation on economic aspects of the protein and energy content of foods (49). In 1887, his series of articles in Century magazine included a description of the energy content of foods that defined the 0–1°C Calorie (kcal) for American audiences (50). Neglecting muscular efficiency, he noted that 1 Calorie equaled ∼1.53 foot-tons of work. It is not evident why Atwater selected the Calorie rather than the g-calorie as used in publications from colleagues in Voit's laboratory.

The articles on nutrition were published the same year that the U.S. Congress enacted the Hatch Act, and Atwater was offered a position as director of the Storrs Agricultural Experiment Station. His interest in establishing a rational basis for food selection led to work on the proximate analysis of foods and their physiological energy value (48). Graham Lusk also visited the Voit laboratory and set the tone for nutrition education in the United States with his text, The Science of Nutrition, which emphasized energetics (39).

Formal definition of the Calorie in the United Kingdom

James P. Joule did not use the calorie as a unit of heat in his published papers until 1878 (29,51). In describing the mechanical equivalent of heat, he monitored the heating of water in °F/pound and expressed work in ft-lb (52). He also discovered the phenomenon of “Joule heating” in which electricity passing through a resistor releases heat and determined the mechanical equivalent of heat in electrical units. Calorimetric studies by Smith (53) measured energy usage in terms of the carbon dioxide formed in metabolism but did not refer to calories. Frankland (54) conducted calorimetry on foods and employed unnamed “heat units” or work units of kg-m (Fig. 2) in the same sense as Mayer (55). Frankland's heat units were identical to g-calories (9) and he stated that his work measured calorific values of foods. He referred to Joule's mechanical equivalent of heat in stating that the heat needed to raise the temperature of 1 kg of water by 1°C equals 423 kg-m of work. In 1865, Fick and Wislicenus observed that protein metabolism accounted for only a minor part of the energy required by muscular exertion during a mountain climb (56). They calculated the energy content of protein (albumen) in heat units that equal modern kcal. In the ensuing international discussion of the study, the kg-m was consistently used as a unit of energy or work rather than the calorie (Fig. 2). Kleiber's book explained that energy expenditure can be estimated by multiplying men's body masses in kg times the vertical distance of the climb and adjusting for mechanical efficiency (57). The usage of kg-m or ft-lbs as units of energy persisted until the joule was adopted as a standard for all energy.

From 1873 to 1913, a committee of the British Association for Advancement of Science met concerning standards for electrical measurements. Among other tasks, the committee solicited international opinion concerning heat units. In 1873, it defined the dyne and erg and recommended that the CGS be adopted (58). In 1888, the joule was named and defined as the work done in 1 s by the power of 1 W (59). The committee also adopted the name “Therm” for “the quantity of heat required to raise one gramme of water at its maximum density by one degree Centigrade” (59). The Therm was discarded in 1896, when the committee discussed whether an international consensus could be determined concerning units of heat (60). A theoretical unit of heat was proposed to be 42 × 10⁶ ergs and a practical unit of heat was proposed to be the heat required to raise the temperature of 1 g of water by 1°C at some temperature between that of ice and 20°C. The report stated:

There is fairly general agreement in the view that as a practical unit the heat required to raise 1 gramme of water 1°C at some fixed temperature must be taken, but views differ as to the temperature which it is most convenient to choose. (60)

In this regard, the committee recommended using the joule, or 10⁷ ergs, as the theoretical unit of heat. It advised that a 10°C thermal unit based on the heating of 1 g of water should equal 4.2 J. The committee indicated that “this secondary thermal unit may be called a Calorie” (60). The word was capitalized although it meant g-calorie, indicating that there was no international convention of capitalizing Calorie to indicate a kcal even though Atwater had used Calorie in that sense in 1887 (50) and his early compendia (48,61,62). In their calorimetric work, Atwater and Rosa (63) defined the Calorie at 20°C and noted that it was slightly larger than the 18°C unit employed by Armsby (64).

Conclusions

This review raises several questions that should be answerable. First, Marcellin Berthelot defined a large calorie as 1000 g-cal (23) but did not call it a kilocalorie. A review of publications in engineering, chemistry, and physics between 1877 and 1894 should disclose a source for the name. Second, one does not know why Voit adopted the g-calorie in 1866 or why Atwater decided to use the Calorie (kcal) after visiting the Munich laboratory. It is likely that both heat units were being used in research publications and engineering handbooks. A review of French and German sources would clarify this question. Third, Nicholas Clément may have been first to coin and define the calorie, but it is possible that the term was already being used by other engineers and chemists. New evidence could falsify the hypothesis that Clément was first. In a larger context, the calorie made its way from France into international scientific literature during the 19th century and important work that is not mentioned here was being done in many other countries.

The Calorie began to enter popular American vocabulary after Atwater explained the unit in his 1887 article in Century magazine. The most important avenue was probably the USDA Farmers' Bulletins (61,62), which provided the first U.S. food databases to be used in dietetics. Then, as now, American audiences were interested in managing weight, and the Calorie was soon introduced in articles and books. For example, Dr. Lulu Hunt Peters' best-selling “Diet and Health with Key to the Calories” specifically cited Farmers' Bulletin 142 as a source of information (65). Eventually, the Calorie was adopted for the nutrition facts panels on U.S. food labels. At present, there does not seem to be a movement by policy makers in the US to replace the Calorie with the kJ on nutrition information panels.

J.L.H. thanks Dr. Patrick Reidenbaugh (University of Georgia Libraries), the staff of the BIPM, the Bibliothèque Nationale de France, and the Library of Congress for bibliographic assistance. Thanks also to Pat Naughtin of Geelong, Australia, for correspondence about the history of the Calorie.