VJhS MCZ library JUN ! 4 2011 Volume 97 HARVARD Number 1 UNIVERSITY Spring 2011 Journal of the WASHINGTON ACADEMY OF SCIENCES Editor’s Comments J. Maffucci i Instructions to Authors Hi Affiliated Institutions Iv Grains, Pulses, and Olives.. .Diet in Ancient Rome M. Brown 1 The Cosmic Microwave Background S. Howard. 25 Genders and International Collaborations L. FrehiH and K. Zippe! 49 Minutes; Science is Murder R. Hietala 71 Membership Application 83 ISSN 0043-0439 Issued Quarterly at Washington DC Washington Academy of Sciences Founded in 1898 Board of Managers Elected Officers President Mark Holland President Elect Gerard Christman Treasurer Larry Millstein Secretary James Cole Vice President, Administration Lisa Frehill Vice President, Membership Sethanne Howard Vice President, Junior Academy Dick Davies Vice President, Affiliated Societies E. Eugene Williams Members at Large Denise Ingram Terrell Erickson Frank Haig, S.J. Alianna Maren Daryl Chubin Michael Cohen Past President: Kiki Ikossi Affiliated Society Delegates: Shown on back cover Editor of the Journal Jacqueline Maffucci Associate Editor: Sethanne Howard Academy Office Washington Academy of Sciences 6^" Floor 1200 New York Ave NW Washington, DC 20005 Phone: 202/326-8975 The Journal of the Washington Academy of Sciences The Journal \s the official organ of the Academy. It publishes articles on science policy, the history of science, critical reviews, original science research, proceedings of scholarly meetings of its Affiliated Societies, and other items of interest to its members. It is published quarterly. The last issue of the year contains a directory of the current membership of the Academy. Subscription Rates Members, fellows, and life members in good standing receive the Journal free of charge. Subscriptions are available on a calendar year basis, payable in advance. Payment must be made in U.S. currency at the following rates. US and Canada $30.00 Other Countries $35.00 Single Copies (when available) $15.00 Claims for Missing Issues Claims must be received within 65 days of mailing. Claims will not be allowed if non- delivery was the result of failure to notify the Academy of a change of address. Notification of Change of Address Address changes should be sent promptly to the Academy Office. Notification should contain both old and new addresses and zip codes. POSTMASTER: Send address changes to WAS, 6*^ Floor, 1200 New York Ave. NW Washington, DC. 20005 Journal of the Washington Academy of Sciences (ISSN 0043-0439) Published by the Washington Academy of Sciences 202/326-8975 email: was@washacadsci.org website: www.washacadsci.org MCZ ^iRRARV i Editor’s Comments Having been editor for this Journal for over a year now, one thing that I can’t get over is the diversity of topics that cross my desk. As a scientist with a specific specialty field, I find that it’s often easy to take for granted how ubiquitous science is. But it is exactly for this reason that I was first attracted to science. Recently I’ve been seeing more advertisements for citizen science projects. I have to say that I love this idea. What better way to get the public involved and interested in science by making it accessible and asking for their participation?!?! This is particularly important to engage the younger generations. As the summer months approach, I challenge all of our members to try to engage the public, and particularly younger students, in science and research. This can be accomplished through informal talks, blog posts, newspaper articles, mentoring programs, science fairs, and the list goes on. If you are looking for some good citizen science projects, I recommend referencing www.scienceforcitizens.net. This is a fairly comprehensive list of research projects that are recruiting citizens to assist. This could be a great way to engage students during their summer break. I encourage you to get creative in engaging your students, family members, friends, and neighbors. This is an opportunity to share the passion that we all have for the science world. This issue of the Journal is a great demonstration of the diverse nature of scientific research. The first article. Grain, Pulses, and Olives: An Attempt toward a Quantitative Approach to Diet in Ancient Rome, authored by Madeline Brown, takes a comprehensive approach to determining the diets of Romans in Ancient Antiquity. Ms. Brown conducted this research during a ten week internship at the National Museum of Natural History following her graduation from Brown University. Following this is an article by Sethanne Howard exploring those ‘wrinkles in space time’ referred to as the Cosmic Microwave Background. The Cosmic Microwave Background, Songs in the Universe explains CMB experiments and what they tell us about how the universe came to be. Lisa Frehill and Kathrin Zippel then present Gender and International Collaborations of Academic Scientists and Engineers: Findings from the Survey of Doctorate Recipients, 2006, where they focus on gender differences among doctoral scientists and engineers when examining the extent to which they collaborate internationally. Finally, we present the minutes from the WAS Spring 2011 2"^ annual Science is Murder event, during which authors Lawrence Goldstone, Ellen Crosby, Louis Bayard, and Dana Cameron talked about their use of science and incorporation of research into their murder mysteries. Enjoy! Jacqueline Maffucci, PhD Editor, Journal of the Washington Academy of Sciences Washington Academy of Sciences Ill INSTRUCTIONS TO AUTHORS 1. Manuscripts should be in Word (Office 03/07) and not PDF. 2. They should be 6,000 words or fewer (exceptions may be made by the Editor). If there are 7 or more graphics, reduce the number of words. 3. Graphics (photographs, drawings, figures, tables) must be in graytone only (no color accepted), and be easily resizable by the editors to fit the Journal’s page size. Do not wrap text around the graphics. 4. References (and bibliography, if included) may be in the format generally acceptable for the disciplinary or professional field represented by the manuscript. They must be accurate, complete, and consistent in format throughout the paper. 5. Include both an e-mail address and a postal address for the author (or primary author) including title and institutional affiliation if any. 6. Papers are peer reviewed. 7. Send Manuscripts by e-mail as an attachment, or on a CD, to Joumal'g^washacadsci.org or directly to the editor. Dr Jacqueline Maffucci - jamaffucci^; gmail.com. Hard copy cannot be accepted. Manuscripts can be accepted by any of the Board of Discipline Editors. Emanuela Appetiti - anthropology at eappetiti^f hotmail.com Elizabeth Corona - systems science at elizabethcorona q gmail.com Jim Eigenreider - science education at iim^g;deepwater.org Terrell Erickson - environmental natural sciences at teiTell.erickson 1 '^a^wdc. nsda.gov Mark Holland - botany at maholland@salisbun .edu Kiki Ikossi - engineering at ikossi@ieee.org Carol Lacampagne - mathematics at clacampagne@earthlink.net Raj Madhaven - engineering at rai.madhaven@nist.gov Kent Miller - computer sciences at kent.l.miller@alumni.cmu.edu Jean Mielczarek - physics and biology at mielczar@phvsics.gmu.edu Robin Stombler - health at rstombler@auburnstrat.com Alain Touwaide - history of medicine at atouwaide@ hotmail.com Steve Tracton - atmospheric studies at straction@hotmail.com Spring 201 1 AFFILIATED INSTITUTIONS The National Institute For Standards and Technology Meadowlark Botanical Gardens The John W. Kluge Center of the Library of Congress Potomac Overlook Regional Park Koshland Science Museum American Registry of Pathology Living Oceans Foundation Washington Academy of Sciences Grain, Pulses and Olives: An Attempt toward a Quantitative Approach to Diet in Ancient Rome 1 Madeline Brown Smithsonian Institution, NHRE Summer 2010 Intern Forward The paper that follows is the first article by Madeline Brown. In May 2010, she obtained a degree in Anthropology from Brown University and, exactly two days later, she was at the National Museum of Natural History of the Smithsonian Institution. With seventeen other college students, she had been selected for a ten week NHRE Internship (Natural History Research Experience Internship). Based on her major in Botany and Ethnobotany, and her interest in Classical Culture (including a class on Roman food), Madeline Brown was directed to my unit, which specializes in the study of medicine, botany, and medicinal plants in the Mediterranean world from the most remote antiquity to the dawn of modem science. I suggested that she explore Roman diet as a possible source of the Mediterranean alimentary tradition. The present article is a presentation of her research, which was entitled “What did ancient Romans eat, and why?” This is the first essay by a freshly graduated student who will probably become a member of the future scientific community. It results from ten weeks of hard work, often late in the evening in the empty US National Herbarium, taking advantage of the Historia Plantarum collection and the documentation and knowledge accumulated in the Institute for the Preservation of Medical Traditions. However limited it will probably be considered, this essay opens the way for future research and is the first announcement of a scientist in-the- making. I am grateful to the Washington Academy of Sciences and the Smithsonian Institution, and also to the Institute for the Preservation of Medical Traditions, for opening their doors to the next generation of scientists and offering them an opportunity to communicate their work, their ideas, and their enthusiasm for the scientific enterprise. Alain Touwaide Smithsonian Institution Spring 201 1 2 Introduction The dietary habits of Romans in Classical Antiquity have been discussed and qualitatively reconstructed in a number of previous studies, but none of these prior efforts have approached both its nutritional and biological properties in a comprehensive or systematic way. In addition, modem dieticians and scholars alike have been interested in the contemporary Mediterranean diet ever since Ancel Keys began his landmark research in the 1950s on the potential health benefits experienced by those who eat traditional Mediterranean diets (Keys, 1970). Yet despite popular interest in this “traditional” Mediterranean diet, little work has been conducted that attempts to uncover the tme origins or the cultural, biological, and nutritional properties of this professed traditional diet. The aforementioned paucity of nutritional and biological quantification in previous studies on the ancient Roman diet may, in part, be responsible for the fact that few connections have been made between the diets of the ancient Romans and those of contemporary Mediterranean people. This pilot study suggests that research examining the actual biological and nutritional properties of the ancient Roman diet may enable us to better understand both the origins of the contemporary Mediterranean diet as well as how this diet has changed over the past two millennia. In this study, the ancient Roman diet is defined as the range of foodstuffs likely eaten in the Mediterranean region from the second century BC up through the fifth century AD. These ranges of time, region, and foodstuffs have been determined by the time period, geographic coverage, and food items that can be traced through the seven primary texts referred to throughout this study. In addition, while the Roman Empire spanned a wide range of areas in Europe, Africa, and the Middle East, this pilot study focuses on the Mediterranean parts of the Roman Empire, specifically Rome and greater Italy. Regional variations in the types and quantities of available foodstuffs undoubtedly occurred throughout the Roman Empire and warrant examination by further studies on the quantification of ancient nutrition. Prior studies on the nutritional properties of the ancient Roman diet have tended to focus on the few food groups, such as cereals and vegetable oils, for which there is fairly concrete textual evidence regarding the extent of their dietary contributions. Gamsey (1998) has made perhaps the most thorough attempt thus far to quantify the caloric and nutritional properties of Roman diets. Based on classical literary references regarding Washington Academy of Sciences 3 the different amounts of grain allotted to various members of Roman society, he calculated the caloric contribution of cereals to the diets of members of each strata of Roman society (Gamsey, 1998). For example, Gamsey concludes that the approximately 33 kg, or five modii (modius being an ancient Roman unit of measure, with one modius equal to about 6.6 kg) of grain per person per month provided under the Roman state’s frumentatio (the allotment of grain provided to Roman citizens by the state) during the 1^^ century AD would have been enough grain to provide about 3,700 kcals per person per day. Gamsey (1998: 229-230) notes that this amounts to almost twice the daily energy needs of an average human being. It must be noted however, that much of this grain would have been both inedible (due to prolonged storage and slow distribution) and contaminated with rocks, dirt or other heavy debris, which would have to be removed from the grain during the winnowing and cleaning process before it could be consumed. Therefore, it can be assumed that ancient Roman citizens receiving the frumentatio would not have actually had access to the full amount of grain (and therefore calories) in their 33 kg allotment. In addition, Gamsey (1998: 236-237) found that Cato’s recommended three modii (c. 19.8 kg) of grain per month for shepherds and estate domestic staff would provide c. 2,200 calories per day, while the four modii (c. 26.4 kg) for agricultural laborers (or slaves) provide c. 2,960 kcals per day. In another notable study, Foxhall and Forbes (1982) suggest that the ancient Romans obtained around 75% of their daily caloric needs from cereals. Other scholars maintain however, that this calculation is merely an estimate, and probably a high one (Gamsey, 1998; see also Schneider, 2006: 916). While these previous studies provide useful frameworks for thinking about the role of grain as a source of food energy for the Roman people, none of them take their quantification methods further in an effort to understand the nutritional properties of the Roman diet as a whole. This study attempts to reconstruct the diet of early imperial period Romans, that is, the inhabitants of Rome and Italy from around the second century BC to the fifth century AD, by expanding on the quantitative methods from previous studies on cereal consumption in ancient Rome and instead quantifying not only the nutritional properties of its cereal components, but also the nutritional properties of the Roman diet as a whole. This is accomplished by determining both the individual and combined nutritional properties of all of the foods that were likely part of the Roman menu. Necessarily, this research relies on literary, archaeological, and botanical Spring 201 1 4 data in order to both develop a comprehensive list of plant and animal species that are likely to have been eaten by the ancient Romans as well as to determine how these specific foods contributed to the overall health of the Roman people. Moreover, by combining modem Mediterranean dietary data with what little is known about consumption patterns in ancient Rome, this pilot study employs a more comprehensive method of quantifying the general nutritional properties of ancient diets than has been used in the past. The results of this analysis suggest that the high quantities of cereals (namely wheat and barley) eaten by ancient Romans may have been sufficient enough to provide the majority of their nutritive needs, with the exception of vitamins A, C, and D, which they instead must have obtained from fruits, vegetables, and exposure to the sun. This pilot study provides a model for one method of quantifying the nutritional and biological properties of the Roman diets and will hopefully help begin to lay the groundwork for introducing further quantification and more holistic evaluation methods into studies on a variety of ancient diets. Methods Primary Sources This study relies on a review of both primary sources and secondary literature discussing the Roman diet, from which a preliminary list of 321 different foods that the ancient Romans likely consumed was developed. Seven primary sources were surveyed to obtain quantitative data on Roman literary references to food (see Table 1). While this cross- section of Roman literature covers only a small percentage of the total classical texts available, it represents a useful and informative selection that is ideally situated for this initial quantitative investigation into the Roman diet. These texts were selected for their focus on ancient Roman food, agriculture, and dietetics, with the acknowledgment that they are a limited selection and of the potential for expansion in both the number and types of literary and other (archaeological, inscriptions, artistic) lines of evidence in future studies. This study drew methodological inspiration from Alain Touwaide’s database of medicinal plants found in classical literature. Following in the methodological footsteps of the Touwaide database, this study expanded upon the preexisting database by determining and focusing on the various food plants and animals (rather than strictly medicinal plants) that are mentioned by ancient Roman authors. Specifically, I referred to the indices and concordances for each Washington Academy of Sciences 5 of these texts (both in their original languages and translation for clarity) and documented how frequently the texts mentioned each of the 321 food items that were likely eaten by the ancient Romans (Briggs, 1983; Dioscorides, 2005; Striegan-Keuntje, 1992). Table 1: Primary Sources Surveyed Period/Date Author Title Edition 234- 149 BC Cato De agricultura Hooper and Ash 1934 116-27BC Varro Rerum rusticarum librilll Hooper and Ash 1934 century AD Columella De re rustica Rodgers 2010 23/24 - 79 AD Pliny Naturalis historia Rackham, Jones, and Eichholz 1938- 1962 4^^ century AD Apicius Apicius Andre 1987 4th _ ^th AD Palladius Opus agriculturae Rodgers 1975 511 AD Anthimus De observatione ciborum ad Theodoricum regim Francorum e pis tula Paolucci 2003 In addition, numerous secondary literature texts were surveyed in order to gain a more informed perspective on the state of research on the dietary properties of the ancient Roman diet (Couplan, 1994; Toynbee, 1973). Several particularly informative texts from this survey are highlighted in Table 2. In order to accurately analyze the number of times each ancient author mentions a particular food source, the total number of food names analyzed in the dataset corresponds to. the total number of unique Latin or Greek food names, rather than to the number of different biological species that these names may represent. This prevents the over- representation of certain foods such as acorns, for which there are four possible species, but only a couple of more generic names in both Latin {glans) and Greek (balanos or drus), which may refer to any of the four different oak species. Therefore, the number ol times that the word glans appears in the primary sources is recorded as only a single data point (for “acorns”) rather than four (one for each oak species) in order to prevent an overrepresentation of acorns in the literature analysis. Spring 201 1 6 Table 2: Notable Contributions from Secondary Literature Date Author Title Brief Description 1947 Thompson A Glossary of Greek Fishes Examination and identification of the fishes mentioned in Greek texts. 1981 Andre L 'alimentation et la cuisine a Rome Overview of Roman cuisine and eating habits organized by culinary categories such as fruits, vegetables, meats, etc. Provided cultural context of Roman dining as well as framework for forming initial list of Roman foodstuffs. 1985 Andre Le noms de plantes dans la Rome antique Scientific and Latin identifications of plants utilized in ancient Rome. 2002 Jashemski and Meyer The Natural History of Pompeii An archaeological examination of organic materials from Pompeii. Provided some archaeological and artistic verification of the identities of foods in Roman diet. 2003 Dalby Food in the Ancient World: From A to Z Encyclopedia of culinary terms, foodstuffs, and other entries related to food and drink in the ancient world. Crucial in helping form initial list of possible Roman foodstuffs. Biological Species Verification Floras of Italy, botanical texts and archaeological evidence from Pompeii were used to identify the plant species found in the Roman diet according to contemporary Linnaean taxonomy (Pignatti, 1982; Van Wyk, 2005; Jashemski, 2002). Determining the scientific names of the otherwise generically named food items was a crucial step in ensuring that the nutritional data used in this study were drawn from geographically appropriate species whenever possible. In addition, verifying the biological identities of foods in the Roman diet allowed us to determine where the plant food items would have originated. Binomial designations ,of plants have been verified for accuracy using the Flora Europaea (Tutin etal, 1980). Washingt.on Academy of Sciences 7 Nutritional Information Quantitative nutritional data for the plant food items included in this analysis were gathered from the “USDA National Nutrient Database for Standard Reference,” which draws its information from the “Nutrient Data Laboratory” of the USDA’s Agricultural Research Service. Nutritional information on the amount of calories, iron, sodium, protein, fat, vitamin A, vitamin D, vitamin C, calcium and sugar per 100 g was compiled for 309 different foods that were likely eaten by the ancient Romans. In gathering these data however, it was not always possible to find the exact nutritional information for each species. In these cases, nutritional data from a closely related species was included instead as a substitute for the missing information according to a procedure that was also frequently used in antiquity (that is, substituting one species for another based on availability). For instance, since there is no publicly available quantitative nutritional data for wild radish {Raphanus raphanistrum L.), the nutritional data from the closely related cultivated radish species {Raphanus sativus L.) was substituted as a nutritional analog. In general, most substitutions of plant or animal species were limited to other species within the same genus (or in some cases, the same family) as the desired species. In those instances when no nutritional information was available for all closely related species of a given food, data from a species with similar growth habits, exploited anatomical parts, or secondary metabolites to those of the species with missing nutritional data were inserted instead. This substitution method was employed primarily in constructing the original dataset, and does not alter the presentation of this study’s results. Nutritional information for the meats was taken from the USDA data on those meat products that most closely resemble the form in which the meat came off of the animal. For instance, the USDA’s nutritional data for “Pork, fresh, carcass, separable lean and fat, raw” was chosen over other potential nutrient datasets such as “Pork, cured, ham, center slice, country-style, separable lean only, raw” or “Pork, ground, 96% lean / 4% fat, raw,” as the nutritional information from a less processed meat product is more likely to resemble the nutritional profile of the meats that were available to the ancient Romans. In addition, grass-fed or wild varieties of animals and plants were used whenever possible in an attempt to further mimic the plant and animal varieties found in ancient Rome, which were undoubtedly both less Spring 201 1 8 domesticated and less selectively genetically engineered compared to most of our contemporary commodity crops and livestock. Foods that are composed of a variety of ingredients (such as cake, bread, beer and aphye) are not included in the nutritional analysis because the exact ingredients, processing methods, and proportions of ingredients for each ancient recipe are currently unknown, and therefore the more general nutritional properties of these food items are also largely unknown. In , addition, wine was not considered in this study, as the classical texts give no clear indication of its frequency of consumption by the ancient Romans, and therefore no indication of its possible nutritional role. Future studies expanding upon this initial quantitative analysis of the Roman diet could significantly improve our understanding of ancient Roman nutrition by considering the dietary role of these additional processed foods such as beer, wine, cakes, and garum. Nutrition Guidelines The nutritional analysis in this study focuses on 10 of the most important and essential nutrients required for proper human nutrition: calories/energy, protein, fat, sugars, calcium, iron, sodium, vitamin C, vitamin A, and vitamin D. This study referred to the World Health Organization (WHO) and the Food and Agriculture Organization (FAO) of the United Nations’ joint recommended daily intake data for vitamins and minerals as the baseline recommended daily amount for the following nutrients: calcium, iron, vitamin C, vitamin A, and vitamin D (WHO and FAO, 2004). In order to compare the Roman diet’s nutritional adequacy for both men and women, our analysis included information on the daily intake required for both adult women (19-50) and adult men (19-65). In calculating the daily recommended amount of iron based on the WHO’s data, the recommended amounts for the middle values of iron bioavailability (12% and 10%) were averaged, as information on the bioavailability of the types of iron in the Roman food sources was unavailable. Besides using FAO and WHO guidelines, this study referred to the recommended daily intake values for protein, fat, sugars, sodium, and calories given by the Confederation of the Food and Drink Industries of the European Union’s (CIAA) Guideline Daily Amounts (GDA) (2010). Washington Academy of Sciences 9 Nutritional Analysis of Diet Because definitive quantitative data on the amount of food consumed by ancient Romans is unavailable, studies on Roman nutrition must instead rely on the information that known, such as the amount of grain provisioned to slaves or soldiers, and relate it to data about contemporary food consumption patterns. Along these lines, this study analyzed the ancient Roman diet using data from the FAO about the amount and proportions of foodstuffs annually consumed by contemporary Mediterranean people. Because the FAO divides diets into categories such as cereals, meats, and vegetable oils, it was possible to estimate the relative amount of nutrients that ancient Romans would have gained from each of these food groups by inserting food items from ancient diets rather than those from modem diets into the appropriate culinary categories. Specifically, this study referred to the FAO’s data regarding the total amount of food available for consumption in Spain, Italy, Greece, and Turkey during the year 1961 (which is the earliest year in which the FAO collected food consumption data). This information was further broken down into the total amount of cereals, starchy roots, sugar crops, sugar and sweeteners, pulses, tree nuts, oil crops, vegetable oils, vegetables, fmits, stimulants, spices, alcoholic beverages, meat, offal, animal fats, eggs, milk, fish and seafood, other aquatic products, and miscellaneous foods. The total kilograms of food available per person per year for each of these countries was then averaged in order to get a general idea of the amount and types of foods consumed by people in the greater Mediterranean. This contemporary information on food consumption provides a baseline idea of the total quantity of food consumed by a single person in a year, as well as how much of each type of food (cereal, meat, fruit, etc.) people in the Mediterranean consume relative to other types of food. There are several key problems inherent in analyzing ancient diets using modem dietary data. While human diets exhibit strong cultural traditions and resilience in their basic components, they also are known to rapidly change to incorporate new foodstuffs made possible by new technology, trade, or other factors. In addition, during the last century, and particularly the last fifty years, human diets have undergone remarkably drastic changes as a result of the increased industrialization of food production and innovative food technologies, as well as unprecedented levels of global trade in agricultural commodities. Therefore, few people today continue to eat the “traditional” foodstuffs that they did several Spring 201 1 10 hundred years or even several decades ago. A second difficulty arises from the differences between the dietary emphases of different food categories (such as meat, dairy, fruits, etc.) of modem and ancient peoples. For instance, contemporary people tend to eat more meat and sugar than those of the past, as unlike in ancient times, both of these food items are now generally no longer expensive, rare, or considered to be luxury foods. These considerations had to be taken into account while conducting this analysis of ancient nutrition using contemporary data about consumption levels. Developing a Model Diet: Grain Consumption as a Baseline Fortunately, the dietary shifts that have occurred since antiquity can be corrected for using the few known quantities of foods that were likely consumed by the ancient Romans, such as the amount of grain. The baseline amount of cereals consumed in the modem Mediterranean diet (according to FAO statistic) was adjusted in order to match the levels of ancient Roman grain consumption suggested by previous studies (Schneider, 2006; Gamsey, 1998; Foxhall and Forbes, 1982). As grain consumption is the only food category with clear ancient textual evidence of its consumption levels, this known quantity of ancient grain consumption provides the foundation upon which this study builds the rest of its nutritional reconstruction of the ancient Roman diet. Based on past analyses of the Roman diet using Cato’s aforementioned writings (Hooper and Ash, 1934) on grain allotments, it has been suggested that the amount of grain allocated to a typical Roman soldier or slave probably fell somewhere between 230 - 330 kg per year (Schneider, 2006: 916). These known quantities of high and low limits of grain consumption from antiquity were incorporated into the contemporary food-intake dataset as the high and low estimates of the relative contribution of grains to the Roman diet. While these amounts of grain may seem high compared to modem levels of consumption, they must be considered within the context of the heavily grain-based ancient diet. According to Pearson (1997:15), during the later Carolingian era, one Anglo-Saxon guideline suggested providing 1.5 - 2 kg of bread per day, as well as additional meat and veggies. These amounts of grain rations are far higher than the intake amounts suggested by classical literature (Cato; Varro; Columella; Pliny; Apicius; Palladius; and Anthimus) and yet, there is clear documentation from throughout the early Middle Ages of a variety of monastic and lay rations of bread Washington Academy of Sciences 11 allotments ranging from 330 g per day up to 1,700 g per day (Pearson, 1997). These medieval bread rations may help to better contextualize the low and high amounts of ancient Roman grain consumption suggested in this study. Two-hundred and thirty kilograms of wheat per year averages out to about 630 g per day, while 330 kg per year averages out to 870 g per day. According to the USDA Nutrient Data Laboratory, 100 g of durum wheat contains 339 kcal. Therefore, these high and low amounts of grain consumption would provide between about 2136 kcal and 2929 kcal per day, which are acceptable amounts of caloric intake for active males relying on a cereal-based diet. Since the ancient Roman diet contained a higher proportion of cereal than the modem Mediterranean diet, this substitution increases, and thereby skews the total mass of food consumed per person per year. To correct for this, the amounts of all non-cereal food items were decreased proportionally so that the total kilograms of food consumed per year remained consistent with the original pre-adjustment amounts. In addition, after making this correction, the proportions of all non-cereal foods in the model diet remained the same as in the original pre-corrected model. This created a nutritional model for the ancient Roman diet based on a realistic amount of total food consumed per year in the contemporary Mediterranean diet. In addition, this method conserved realistic proportions between the different types of non-cereal foods while also enabling the simulation of a cereal-heavy diet that more closely resembles that of the ancient Romans (rather than the modem Mediterranean people). An analysis of our results after these initial calculations revealed that the introduction of higher amounts of cereals (which are calorie-dense foods) actually altered the dataset to be unrealistically high in calories. To adjust for this, the total amount of food consumed per person per year was recalculated, this time including only those items that are known from primary sources and archaeological evidence to have been commonly consumed in the ancient Roman diet. This left us with a dataset containing information on the annual consumption of the ‘‘core” food categories in ancient Roman diets, that is, cereals, pulses, tree nuts, oil crops (olives, sesame seeds, etc.), vegetable oil, vegetables and starchy roots, fish and seafood, spices, and eggs. Notably, milk, meat and fruit were taken out of the “core” dataset, as the classical texts (Cato; Varro; Columella; Pliny; Apicius; Palladius; and Anthimus) suggest that the ancient Romans seem to have not consumed these foods other than on an irregular or seasonal basis. In addition, while it is well known that all Roman citizens, regardless of class, had access to meat during public festivals throughout Spring 201 1 12 the year, it is not well known exactly how many animals were slaughtered or how the meat was distributed at each of these festivals, and therefore what the nutritional role of meat would have been for the Roman people. Therefore, this initial study does not consider meat as being a significant part of the Roman diet, as it is yet unclear how regularly or in what quantities meat was actually consumed. Once the core categories of food in the Roman diet were established, the amount of food likely consumed by a Roman person from each of these categories in an average year was calculated. By combining this information (on total mass of food consumed) with information on the average nutritional properties (that is the amount of calcium, fat, etc. per 100 grams) of each food, it was possible to create a general model of the dietary profiles of people who consume the types of foods that the Romans ate in the same relative amounts and proportions (as determined using the previously discussed model). In addition, the high and low bounds of grain consumption also provide analogs for the nutritional profiles of both the lower and middle-to-upper classes in Roman society, as lower class individuals would have relied more heavily on grain, while those of upper and middle classes likely enjoyed a more varied diet. Results Foods Mentioned in Primary Sources Based on this survey of seven primary sources that discuss food and agriculture in the Roman Empire, it is clear that ancient Roman authors mentioned some types of food far more frequently than others. This information has been essential for determining and confirming the types of foods eaten by the ancient Romans, and thus the types of foods included in this study’s nutritional analysis. Table 3: Mentions of Unique Foods in Primary Sources (listed chronologically) Classical Author Number of Unique Foods Discussed Cato 63 Varro 86 Columella 141 Pliny 117 Apicius 128 Palladius 136 Anthimus 56 Washington Academy of Sciences 13 The total number of different foods mentioned by the classical Roman authors surveyed in this study varies greatly (see Table 3). Altogether they mention 225 different foods, though as previously mentioned, some of these Greek and Latin names of foods may correspond to multiple biological species. The top ten foods mentioned in the classical texts that were surveyed in this study are indicated in Table 4. The wide variation in the frequency of literary references per food is apparent even in these top ten items. Furthermore, only 26 of the 225 different food items found in the ancient texts are mentioned over 100 times; by contrast 44 different foods are mentioned five times or less. This indicates that while ancient authors clearly had knowledge of a wide repertoire of culinary options, they in fact focused their discussion of food and dietetics on a more limited number of foods. Table 4: Top 10 Food Items Mentioned in Seven Classical Texts Food Item Number of Times Mentioned Black pepper 463 Wine 368 Grapes 305 Honey 266 Vinegar 253 Figs 244 Barley 210 Rue 206 Lovage 189 Olives 181 Nutrition Using both high and low estimates of ancient Roman cereal consumption (230 and 330 kg per year respectively), this study generated two different profiles of the nutritional properties of ancient Roman diets depending on whether one assumes it contained higher or lower proportions of cereal compared to the overall amount of food consumed (Table 5). Assuming the Romans consumed the higher level of wheat intake (330 kg per year), and using the quantitative methods previously outlined, this study found that the total mass of food in the Roman diet following this model would have been 65% cereals, 28% vegetables and starchy roots, 2% fish and seafood, 1.8% vegetable oils, 1.1% pulses, 0.9% eggs, 0.8% tree nuts, 0.4% oil crops, and 0.1% spices. Assuming a lower level of wheat consumption (230 kg per year), these values shift so Spring 201 1 14 that the dietary mass is 50.4% cereals, 39.6% vegetables and starchy roots, 2.8% fish and seafood, 2.5% vegetable oils, 1.6% pulses, 1.2% tree nuts, 1.2% eggs, 0.6% oil crops, and 0.2% spices. Table 5: Amount of Nutrients in Roman Diet Based on Daily ^ecommenc ed Nutrient Quantities (percent ages) Iron v.c Prot. Cal. V . A Fat Calc V. D Salt Sug. M e n Low* 285.8 224.1 170.5 117.2 121 83.2 68.2 42.9 12.4 10 High 357.3 180 217.9 150.1 95.3 82.4 70.4 33.7 10.4 9.2 VV 0 m e n Low 133 224.1 204.6 146.6 145.2 95.1 68.2 42.9 12.4 12.3 High 166.4 180 261.5 187.6 114.3 94.2 70.4 33.7 10.4 11.2 *Low and High designations refer to models created based on the low and high estimates of the contribution of wheat in the Roman diet, 230kg and 330kg per year respectively. Discussion Composition of Roman Diet The Romans primarily ate cereals and legumes, often supplemented with vegetables, cheese, or meat and covered with sauces made out of fermented fish, vinegar, honey, and various herbs and spices (Schneider, 2006: 919). While they had some refrigeration, much of their diet depended on which foods were locally and seasonally available. Meat and fish were luxuries primarily reserved for the upper and upper-middle classes, although lower class Romans sometimes obtained low-quality meat from either public sacrifices or urban cookhouses (Herz, 2006). These results are visually reflected in pyramids (Figures la and lb, see Appendix), which show the relative importance and nutritional contributions of various foods to the ancient Roman diet (the most frequently consumed foods are shown at the bottom of each pyramid, while the least frequently consumed foods are shown near the top). Since meats and fish were more often eaten as luxuries than as everyday foods, the protein from pulses played a fairly significant nutritional role for the ancient Romans. Gamsey (1998) even suggests that lentils, broad beans, and chickpeas provided most of the non-cereal calories and protein for the Roman people. While the classical texts do not clearly indicate the amount of meat commonly consumed in ancient Rome, archaeological evidence suggests Washington Academy of Sciences 15 that Roman people across most socioeconomic classes consistently consumed at least a limited quantity of meat. Cucina and colleagues analyzed 77 skeletons from the Necropolis of Vallerano, a suburb of Rome and found a low frequency of oral pathologies, which they correlate with a diet that includes meat and “low-calorific” crops, but which is “seemingly low in refined carbohydrates” (Cucina et al, 2006: 104). In addition, Cucina and colleagues suggest that based on the presumed social classes of these skeletons, “their diet mainly consisted of cereals and low-cost goods,” while the more valuable foods (such as meat, dairy, spices, etc.) were instead reserved to sell at the nearby markets (Cucina et al, 2006: 106). In conclusion, these authors suggest that the lack of oral pathologies in the skeletons examined at Vallerano is not inconsistent with a primarily vegetarian diet, as long as grains made up a lower proportion of the diet compared to other vegetable foods (Cucina et al, 2006: 115). Results from Cucina and colleagues (2006) rely on osteological data to support the supposition that populations in Roman suburbs, which primarily grew perishable crops rather than grain, would have eaten less grain than their urban and rural counterparts. Although Cucina and colleagues’ (2006) osteological study provides important evidence of the nutritional profiles of one community of Roman people, their results cannot be generalized to explain Roman nutrition as a whole. As this pilot study attempts to quantify the nutritional properties of Roman diets in general, it relies less on specific osteological case studies and instead more on both classical literature and contemporary nutritional information. Future studies on ancient nutrition should incorporate both archaeological and osteological evidence in order to more clearly examine how the nutritional profiles of ancient Romans described in classical texts differs from those suggested by alternative lines of evidence. In addition to the staple foodstuffs of grains, pulses, and occasionally meats, the ancient Romans also enjoyed a wide variety of fruits, vegetables, and exotic spices and condiments. While fresh fruits were probably only seasonally available in ancient Rome, Columella speaks to the important nutritional role of dried fruits (such as apples, figs, and pears) in rural people’s winter diets (Columella Book 12, XIV; noted in Brothwell and Brothwell, 1998: 146). These dried fruits likely provided both essential vitamins and small amounts of sugar for the Roman people who had access to them. Contrary to Schneider’s (2006: 918) assertion that Cato provided his slaves with both figs and preserved olives, Cato does not in fact mention that figs were given to his slaves. Instead, he clearly discusses providing cereals and preserved olives to his slaves (Cato Spring 201 1 16 56, 58), and limits his remarks on figs to methods of cultivation, harvesting, and preserving them,, rather than to their role in the diets of different members of Roman society (see esp. Cato 99, 143, 8.1, 94). Cato leaves us unsure about whether or not ancient Roman landowners provided figs for their slaves. While fruits may have been nutritionally important to certain classes of ancient Roman people, the results of this study’s classical literature review support Schneider’s (2006: 918) assertion that neither wild animals nor wild plants are likely to have played large roles in the nutritional status of the ancient Romans. Cultivated tree nuts however, may have played a small but significant role in the Roman diet. This is because while they were usually only eaten as condiments or as dessert, they are exceptionally high in calories, fat, and protein, meaning that they can still affect human nutrition even when consumed in small amounts (Brothwell and Brothwell 1998: 149; USDA Nutrition Database). A variety of sodium-filled condiments may also have had a significant impact on ancient Roman nutrition. The ancient Romans seem to have enjoyed covering their foods in complex sauces, made with many different ingredients and oftentimes possessing fairly distinct flavors. One sauce for oysters and shellfish from Apicius for example, calls for “pepper, lovage, parsley, dry mint, bay leaf, malabathrum [leaves of Cinnamomum tamala], plenty of cumin, honey, vinegar, and liquamen [fermented fish sauce]” (Brothwell and Brothwell 1998: 66). In general however, the main sauce enjoyed by the ancient Romans was garum, a fermented and salted fish sauce, which they applied liberally to savory and sweet dishes alike. Smriga and colleagues (2010) analyzed the nutritional properties of garum found in residues left on pots from the “Garum Shop” in Pompeii. They found that the ancient garum residue contained amino acids in amounts comparable to those found in modem Southeast Asian and southern Italian fish sauces, with free glutamate, glycine, and alanine providing most of the flavor and amino acid content of the sauce (Smriga et al, 2010: 442). While the high levels of amino acids and salt in gamm undeniably played an important role in Roman nutrition and digestive processes, classical texts do not clearly discuss the quantities in which garum was consumed by ancient Roman peoples, which therefore prevents us from quantifying its nutritional role in this initial study. Washington Academy of Sciences 17 Nutrition This study outlines nutritional properties of a Roman diet in which individuals maintain a continually high level of consumption of both cereals and supplementary non-cereal foods. It is unlikely however, that all (or even most) Romans had access to this wide variety of food items in such quantities at all times of year (Schneider 2006: 917). Furthermore, Gamsey (1998: 240) suggests that while most lower class Romans could expect to supplement their grains with some poor quality wine, legumes and olive oil, and even sometimes with vegetables, fish, or fish-sauce, other animal products such as meat, eggs and dairy were rarely obtainable for these people. In addition, while Table 6 includes information on the nutritional intake for both men and women, these numbers assume that men and women had access to the same dietary quality and quantities. Due to various cultural considerations and ancient medical theories however, it is likely that ancient Roman women would not have eaten as much food as men on a daily basis. Finally, this study does not take longevity into consideration as a factor in and byproduct of ancient nutritional intake. Neither lifespan nor differences in nutritional intake across the various phases of life are indicated in the classical texts, and thus the relationship between nutrition and longevity could not be taken into account in this study. Future studies that examine osteological evidence in conjunction with ancient literary sources may be able to better understand and examine these issues. The high proportion of cereals and legumes in the ancient Roman diet provided them with substantial amounts of calories, protein, calcium, and iron. Because their staple foods are deficient in vitamins A, C, and D, the Romans likely obtained these nutrients from seasonally available fruits and vegetables, although spices may also have played a role in providing these nutrients. This analysis indicates that their diet was fairly low in vitamin D, sodium, and sugar. It has to be expected however, that the ancient Roman people’s high sun exposure and proximity to the sea also had positive health effects, conferring both vitamin D and iodine respectively. One unanticipated conclusion of this study is that in the quantitative model, pulses make up a much lower percentage of the ancient Roman diet than might otherwise be expected based on qualitative information from primary literary sources. This disparity may be attributable to the fact that concrete evidence regarding the actual amount of legumes consumed by the ancient Romans is largely uncertain and Spring 201 1 18 unavailable. Therefore, this initial study instead relied on the known amount of legumes consumed in modem Mediterranean diets in order to calculate the nutritional role of legumes in the ancient Roman diet. This difference may have created the disparity in this study’s conclusions, as legumes were likely not eaten in the same dietary proportions by both the ancient Romans and modem Mediterranean people. As quantification methods continue to be explored and improved upon in future studies, disparities such as this one seen with legumes will hopefully be more clearly resolved and understood. Despite the aforementioned problems associated with using contemporary food consumption data as a model for ancient diets, this quantitative method of inquiry can be effectively employed to create a general model of the core nutritional profile of ancient Roman diets. The data gathered using this type of quantitative modeling should be considered only while keeping these conceptual issues in mind and carefully considering the available qualitative information regarding both ancient and contemporary diets. Conclusions As has been suggested by previous studies, the overall nutritional properties of the Roman diet are largely dependent on its three main components: grains, wine, and olives (or olive oil) (Schneider, 2006: 919). This study takes our prior knowledge of the three basic components in the Roman diet and broadens our analytical focus to encompass the biological and nutritional properties of all of the possible food items consumed by ancient Romans. This method has not been free of problems however, as other than quantities for the amount of grain that was given to slaves and soldiers per year, there is little quantitative data from ancient Rome on how much of a given food item people tended to consume in a given time period (Gamsey, 1998). While quantitative methods for investigating ancient nutrition may still be somewhat imprecise and rely on various estimates and assumptions about ancient food consumption, they do provide us with a useful overview and general profile of the probable nutritional and biological properties of the Roman diet as a whole. This study found that the core constituents of the Roman diet (cereals and legumes) meet many of the daily nutritional needs of men and women when consumed in the amounts presumably eaten in ancient Rome. The Romans likely met their additional nutritional needs using a wide variety of more sporadically consumed foods such as meats, fruits. Washington Academy of Sciences 19 and spices. The nutritional impacts of meat, wine, garum and other specialty food items in the Roman diet should be further explored in future studies, perhaps by incorporating additional archaeological and osteological evidence. In addition, further examining how ancient diets varied throughout the Roman Empire could perhaps lead us to better understand both the origins and spread of Mediterranean dietary traditions as well as how nutritional profiles may have varied throughout the empire. This investigation also found that neither the nutritional properties nor the frequency of consumption for a given food seem to have been the determining factor for how often it was mentioned by classical authors. This initial finding of a lack of correlation between these variables is intriguing and warrants further investigation into uncovering why the ancient Roman authors mentioned different food items in such varying frequencies. Although we may never be able to know exactly what the ancient Romans ate, we should continue to attempt to quantify and better reconstruct the nutritional properties and composition of their diet through further investigations. This initial study provides one model for introducing such quantification methods into research on ancient diets. 20 References Andre, Jacques. Apicius, L'art culinaire. Texte etabli, traduit et commente - (Collection des Universites de France). Paris: Les Belles Lettres, 1987. Andre, Jacques. L 'alimentation et la cuisine a Rome. Paris: Les Belles Lettres, 1981. Andre, Jacques. Le noms de plantes dans la Rome antique. Paris: Les Belles Lettres, 1985. Briggs, W.W. Jr. Concordantia in Varronis Libros De Re Rustica. Hildesheim: Georg 01ms Verlag, 1983. Brothwell, Don and Patricia Brothwell. Food in Antiquity: A Survey of the Diet of Early Peoples. Expanded edition. Baltimore: Johns Hopkins University Press, 1998. Confederation of the food and drink industries of the EU (CIAA). Guideline Daily Amounts (GDAs) - GDAs Explained, 2010. http://gda.ciaa.eu/asp2/gdas_portions_rationale. asp?doc_id=l 27. Couplan, Francois and Eva Styner. Guide des plantes sauvages comestibles et toxiques. Paris: Delachaux et Niestle, 1994. Cucina, Andrea, Rita Vargiu, Domenico Mancinelli, R. Ricci, Elena Santandrea, Paola Catalano, and Alfredo Coppa. “The Necropolis of Vallerano (Rome, 2”^ - Century AD): An Anthropological Perspective on the Ancient Romans in the Suburbium. in International Journal of Osteoarchaeology 16: 104-117, 2006. Dalby, Andrew. Food in the Ancient World: From A to Z. Eondon: Routledge, 2003. Dioscorides, Pedanius. De materia medica. Translated by Eily Y. Beck. Hildesheim: 01ms - Weidmann, 2005. Food and Agriculture Organization of the United Nations (FAO). FAOSTAT Food Balance Sheets (Italy, Turkey, Greece, and Spain). http://faostat.fao.Org/site/368/default.aspx#ancor. Foxhall, Ein and Forbes, Hamish A. “Sitometreia: the role of grain as a staple food in classical antiquity.” in Chiron 12: 41-90, 1982. Gamsey, Peter. Cities, Peasants and Food in Classical Antiquity: Essays in Social and Economic History. Walter Scheidel, ed. Cambridge: Cambridge University Press, 1998. Herz, Peter. “Meat, consumption of,” in Hubert Cancik and Helmuth Schneider (eds.). Brill's New Pauly. Encyclopedia of the Ancient World, vol. 8. Eeiden and Boston: Brill, 2006, cols. 535-537. Hooper, William Davis. Marcus Porcius Cato, On Agriculture. Marcus Terentius Varro, On Agriculture. With an English Translation. Washington Academy of Sciences 21 Revised by Harrison Boyd Ash (Loeb Classical Library 283). Cambridge MA: Harvard University Press, and London: William Heinemann, 1934. Jashemski, Wilhelmina and Frederick Meyer eds. The Natural History of Pompeii. Cambridge: Cambridge University Press, 2002. Keys, Ancel. Coronary heart disease in seven countries. Circulation 41 (suppl. 1): 1-211, 1970. Paolucci, Paula. Anthimi epistulae de observatione ciborum ad Theodoricum regem Francorum Concordantiae. Hildesheim: 01ms - Weidmann, 2003. Pearson, Kathy. “Nutrition and the Early-Medieval Diet.” in Speculum 72(1): 1-32, 1997. Pignatti, Sandro. Flora dTtalia. Bologna: Edagricole, 1982. Rackham, Harris, William Henry Samuel Jones, and David Eichholz, Pliny, Natural History, with an English Translation 10 vols. (Loeb Classical Library 330, 352-353, 370-371, 392-394, 418-419). Cambridge MA and London: Harvard University Press, 1938- 1962. Rodgers, Robert Howard. Columellae res rustica, Incerti auctoris, Liber de arboribus. Recognovit brevique adnotatione critica instruxit - (Scriptorum classicorum bibliotheca Oxoniensis). Oxonii: E Typographeo Clarendoniano, 2010. Rodgers, Robert Howard. Palladius, Opus agricultiirae, de veterinaria medicina, de insitione. Edidit - (Bibliotheca Scriptorum Graecorum et Latinorum Teubneriana). Leipzig: B. G. Teubner, 1975. Schneider, Helmuth. “Nutrition,” in Hubert Cancik and Helmuth Schneider (eds.), BrilVs New Pauly. Encyclopedia of the Ancient World, vol. 9. Leiden and Boston: Brill, 2006, cols. 914-921. Smriga, Miro, Toshimi Mizukoshi, Daigo Iwahata, Sachise Eto, Hiroshi Miyano, Takeshi Kimura, Robert 1. Curtis. “Amino acids and minerals in ancient remnants of fish sauce (garum) sampled in the “Garum Shop” in Pompeii, Italy.” in Journal of Food Composition and Analysis 23: 442-446, 2010. Striegan-Keuntje, Ilona. Concordantia et Index in Apicium. Hildesheim: 01ms- Weidmann, 1992. Thompson, D’Arcy Wentworth. A Glossary of Greek Fishes. London: Oxford University Press, 1947. Touwaide, Alain. Medicinal Plants of Antiquity. Unpublished computerized database. Spring 2011 22 Toynbee, Jocelyn M.C. Animals in Roman Life and Art. Ithaca, New York: Cornell University Press, .1973. Tutin, Thomas Gaskell, Vernon Hilton Heywood, Norman Alan Burges, David M. Moore, David Henriques Valentine, Stuart Max Walters, and David Allardice Webb. Flora Europaea. Cambridge: Cambridge University Press, 1980. USDA Agricultural Research Service, Nutrient Data Laboratory. USDA National Nutrient Database for Standard Reference. http://www.nal.usda.gov/fnic/foodcomp/search/. Van Wyk, Ben-Erik. Food plants of the world: an illustrated guide. Portland, Oregon: Timber Press, 2005. Varro, Marcus Terentius. On Agriculture, with an English translation by William Davis Hooper, revised by Harrison Boyd Ash, [Loeb Classical Library No. 283] Cambridge MA: Harvard University Press, 1935, pp. 159-529. World Health Organization (WHO) and FAO. Vitamin and mineral requirements in human nutrition: Second edition, 2004. http://whqlibdoc.who.int/publications/2004/9241546123.pdf. Acknowledgments I wish to thank the Smithsonian Institution, National Museum of Natural History, for my selection in the first session of Natural History Research Experience (NHRE) Interns during the summer of 2010. I am grateful to my supervisor. Dr. Alain Touwaide, Historian of Sciences at the National Museum of Natural History, for having suggested to do research on the history of food in antiquity, his scientific direction, ongoing advice, patience, and support during both the research and writing phases of this study. His constant assistance and expertise have helped develop this study into its final form. In addition, Emanuela Appetiti, Scientific Program Specialist at the National Museum of Natural History, contributed to make my stay at the Smithsonian productive and enjoyable, and provided stimulating views on my research, including the writing of this article. Finally, I express my gratitude to the anonymous reviewers for their constructive suggestions and thoughtful comments, which have enabled me to avoid former shortcomings and have helped me better develop some components of an earlier version of this article. As always, any remaining imperfections are mine. Washington Academy of Sciences 23 Appendix: Figure la: Suggested Relative Role of Foods in Roman Diet by Food Category Herbs' '& Spices'^ Tree Nuts 'Muisum Wine Must Honey Vinegar | Meat Eggs '& Dairy Vegetable -ish Oils and Oil Crops Pulses Fruit: Vegetables and Starchy Roots' Cereals Spring 201 1 j 24 Figure lb: Suggested Relative Role of Foods in Roman Diet by Name ;isum Wine '.'ust Hone. V 'Milk Eggs 'Cheese ClMCkan Pifindga' Oalnch CnraOog^ Ouck FMfTWigo ^ HCKM Goat Oatmau*i Qoom FfUKSkn The. tow* Poni Ptgson' r Sen Tl AieiaiKJefs Bfoadteaved Peppetvneed Artichoke Morel Asphodel Beet Black bryony Black morel Black truffle Orchis^ Cachrys kbanobs Cardoon Carrol Ctfs Ear Arum Desert t ChoncMa juncea Costus or putchuk Cress Cucumber Chicory Elecampane Enrtve Enngo Field eryngo Field Mushroom. Garkc ’ Goideh chamarefle Golden thistle Groundsel Horse Mushroom Houtxfs Tongue Kale Leek Lettuce WM lettuce Horseraiksh Lasura ' Lovage Mallow Settle. Onun Wild orach Mountam orache Celery Pvsnp PeWory Poron Purslane Radish Red Truffle Rocket Sainptwe S.*ia«oi Skirret Soapwort Sorrel. Patience Dock Sow Thode Sp*enard Spotted Golden Thistle Teasel Truffle Tum^r Water pepper Watercress Tarol White mustard White Truffle Whitetop WM Radsh Winter Truffle Asparagus Grape Hyacmth Grass Wy MetKerraiean giaaoius Satytlon Cabbage Gourd ^ Durum Wheat Barley Sorghum Einkorn Emmer Wheat 'Millet Oats Purple Amaranth Rice Rye' Washington Academy of Sciences The Cosmic Microwave Background Songs in the Universe 25 Sethanne Howard USNO, retired Abstract The Cosmic Microwave Background gives astronomers a lot of information about our early universe and explains why it has structure. Yet the details are a bit esoteric. Reviewed here are the basic concepts and discoveries of the Cosmic Microwave Background from an astronomer’s point of view. Introduction The Cosmic Microwave Background (CMB) is a hot (so to speak) topic in astronomy. Sometimes called ‘wrinkles in space-time’, the CMB tells us about the birth and evolution of the universe we see. It maps the earliest moments of our universe. Recent CMB experiments have firmly established the Big Bang Model as the leading theory in cosmology. Yet, despite its fascination, the CMB can be tricky to decode. So let us look at some concepts that will help us understand these experiments and what they tell us. We call the electromagnetic radiation that still lingers from the very early universe the cosmic microwave background radiation - cosmic for its origin in the very early universe, microwave because it shows up in the microwave section (1.9 mm) of the electromagnetic spectrum, and background because it fills all of space. Several people had predicted the existence of the CMB as early as the 1940s. In 1941 the chemist Gerhard Herzberg almost discovered it in a stellar spectrum. The CMB’s serendipitous discovery in 1964 by radio astronomers Amo Penzias and Robert Wilson earned them the 1978 Nobel Prize. They were using a radio telescope (antenna) to scan for signals bouncing off echo balloon satellites. They went to great lengths to track down a ubiquitous 1.9 mm ‘noise’ in their data. They even removed the “white dielectric material” left in the antenna horn by nesting pigeons. They also removed the pigeons. These CMB photons are all around us. They are so weak, however, that it takes very sensitive microwave detectors to detect and measure Spring 201 1 26 them. Optical telescopes will not do the job. The space between stars and galaxies (the background) is dark to an optical telescope. But a sufficiently sensitive radio telescope shows a faint background glow, almost exactly the same in all directions, that is not associated with any star, galaxy, or other object. This glow is strongest in the microwave region of the radio spectrum (that 1.9 mm wavelength). The CMB fills the universe and can be detected everywhere we look. In fact, if we could see microwaves, the entire sky would glow with a brightness that is astonishingly uniform in every direction. The temperature (associated with the peak in the energy distribution at 1.9 mm)^ of this background is uniform to better than one part in a hundred thousand (z.e., the temperature variation on different angular patches of the sky is less than 1 part in 100,000). This uniformity is one compelling reason to interpret the radiation as remnant heat from the Big Bang; it is very difficult to imagine a local source of radiation that is this uniform. A perfectly uniform background distribution will not explain why our universe looks the way it does. The universe contains clumps of matter of all sizes from atoms to galaxies. Such rich structure could not form from a perfectly smooth background. We need to have some wrinkles in the otherwise smooth background. If we have small wrinkles (bumps, ripples) or hills and valleys early in the universe, matter will tend to fall into the valleys, eventually producing dense regions that become the sites of galaxies. Matter attracts matter, so these ‘valleys’ get denser, eventually coalescing into galaxies and clusters of galaxies. Figure 1 (an idealization) indicates the concept. Figure 1. The top figure shows conceptual hills and valleys. The bottom figure shows the top view of the same thing where the grey scale coding refers to the density of matter (dark regions have more matter, light regions less). Washington Academy of Sciences 27 As we look at results from CMB experiments, then, we will expect to see the observed smooth distribution and hope for some tiny amount of rippling that will resemble the bottom of Figure 1. To interpret these CMB experiments properly, we need to look at the origin of the CMB. The action begins at the very early universe. The Big Bang Model predicts (trust me on this - the actual theory is really messy) that the CMB originates from a time just a mere 380,000 years after the Big Bang.” Let us just say that everything starts with the Big Bang (of course). The vast majority of astronomers use the Big Bang Model to describe the origin and evolution of our universe. To learn how the CMB came to be, we, like the astronomers, need to start with the Big Bang. The Big Bang The Big Bang Model rests on two theoretical pillars: 1. The General Theory of Relativity. In 1916 Einstein proposed his General Theory of Relativity (GTR) as a new theory of gravity. Gravity became, not the gravitational field of Isaac Newton, but a distortion of space and time itself. Physicist John Wheeler put it well when he said “Matter tells space how to warp, and warped space tells matter how to move.”'” The theory continues to pass a series of ever more rigorous tests. 2. The Cosmological Principle. We assume matter in the universe is homogeneous and isotropic when averaged over very large scales. This assumption is called the Cosmological Principle. It is the simplest assumption to make - that if you viewed the contents of the universe with sufficiently poor vision, they will appear roughly the same everywhere and in every direction. A homogeneous universe contains the same stuff regardless in which direction you look. It is well mixed (aka homogenized milk). The physical conditions are the same at every place. An isotropic universe looks the same regardless of which direction you look. You cannot distinguish one direction from another. If the universe looks fiat over there then it must look flat over here. This assumption is tested continuously as we observe the distribution of galaxies on ever larger scales. When we look at the universe with galaxy sized eyes we see a clumpy universe with galaxies scattered about and clustered into groups. On smaller scales we see individual stars, some that cluster into groups. Spring 201 1 28 and some that stand alone. And, of course, on even smaller scales we see individual people. It is only when we look at the universe as a total system that we can assume homogeneity and isotropy. After the introduction of GTR a number of scientists, including Einstein, applied the new gravitational dynamics to the universe as a whole. In 1927 Georges Lemaitre used GTR to develop the Big Bang Model. This model predicts that the universe, originally in an extremely hot and dense state, has since cooled by expanding to the present diluted state, and continues to expand today. Two years later Edwin Hubble made one of the profound observations of the early 20^ century - the universe is expanding - supporting the Big Bang prediction of an expanding universe.''^ To interpret the expansion properly requires an assumption about how the matter in the universe is distributed, hence the cosmological principle. The cosmological principle and GTR form the basis for Big Bang cosmology and lead to very specific predictions for observable properties of the universe. For example, given the assumption that the matter in the universe is homogeneous and isotropic it can be shown (again, trust me) that the corresponding distortion of space-time (due to the gravitational effects of this matter) can have one of only three forms, shown schematically in Figure 2. It can be positively curved like the surface of a ball and finite in extent; it can be negatively curved like a saddle and infinite in extent; or it can be flat and infinite in extent - our ordinary conception of space. A limitation of Figure 2 is that we can only portray the curvature of a 2- dimensional plane of an actual 3-dimensional space. Note that in a closed universe you could start a journey off in one direction and, if allowed enough time, ultimately return to your starting point; in an infinite universe, you would never return. Which form the universe adopted will await experiments. Of course there is a problem with the cosmological principle - called the horizon problem - the puzzle that the universe looks the same on opposite sides of the sky even though there has not been enough time since the Big Bang for light (or anything else) to signal across the universe and back. Light can only travel so fast. So how do the opposite horizons “know” how to keep in step with each other, to maintain isotropy? In other words, the universe is too big for its horizons. This cosmological problem has a solution. It is called the era of inflation. We shall meet the era of ’ inflation in the next section when we travel backward in time. Washington Academy of Sciences 29 n„