When used as food, the salt is removed by washing or soaking in water. The recovery of fruit fragments just outside some buildings in Berenike suggests local growth of the plant, which would have eliminated the need for preservation. Members of the white bryony family are very juicy in a fresh condition and cannot be dried successfully. Four species are evidenced in Berenike by their seeds: the colocynth , the watermelon , the cucumber , and the bottle gourd . The colocynth grows in the Eastern Desert, where it is easily gathered and remains fresh for at least a couple of weeks. Ripe water melons can only be stored for five up to eight days and must have been available from local kitchen gardens or harvested in an unripe condition if they were imported. The ripening process continues after harvesting if the fruits are exposed to warm air, preferably between 20 to 25°C. Methods of preserving gourds until the next yield are available, is described by Pliny . One way is to add brine and another is to store them in a trench floored with sand and covered with dry hay and earth. In this way, gourds will remain in a green condition. The fruits of the subfamily Maloideae are classified as pomes. The seeds are enclosed in a cartilaginous structure in the center of the soft, indehiscent fruit. Apples can be preserved in several ways: fresh, dried, or kept in a preservative liquid, such as honey, as described by Apicius , or in a mixture of vinegar and brine as mentioned by Columella . Fresh, juicy apples might have been the most desirable ones. Dealing with the preservation of fresh fruits, Varro, cited by Pliny , bud drying rack says that fruits such as apples should be separately wrapped in fig leaves. According to Galen , properly ripened apples can be stored for the winter and the following spring . Windfall specimens were not used for storage.
The storage life of fresh apples is prolonged if stored in a cool environment. Although this condition is not met while on the way to Berenike, the transport of fresh apples to this remote destination would not have been problematic from September to October, just after harvesting. Today, dried apples produced on a commercial scale are peeled, cored, and sometimes cut into rings. But small apples can be dried as whole specimens or simply cut into segments, as described by Columella . In the latter case, fruits evaporate quicker and spoilage is reduced. Such halved apples, preserved by charring, are frequently found in archaeological contexts. The presence of pips in Berenike leaves open all aforementioned kinds of preservation methods. The last group of fruits and nuts to be dealt with concerns false fruits. They are represented by the juniper and the fig . The berrylike fruit of the juniper is easily dried and can be kept for a considerable period. Fresh accessory fruits of the fig are juicier but can also be dried easily. Due to the high sugar content, they can be kept until the next harvest. Long-distance transport is no problem, as is evidenced by the recovery of figs throughout the Roman Empire. Fresh figs can only be kept for a couple of days, as they are vulnerable to mold, which results in lactic fermentation. It is possible to preserve fresh figs in honey, as is mentioned by Apicius . Fig paste, made from trampled figs mixed with sesame , anise , fennel , and cumin , was made into balls and kept in jars covered with pitch or was heated in vessels in order to reduce the moisture content .The archaeobotanical record of Berenike and Shenshef, based on samples from seven excavation seasons, comprises 68 cultivated plants that could be identified in most cases to the level of species. As much as 28 cultivated plant species are evidenced from Berenike but are absent from the archaeobotanical record from Shenshef. Juniper is, on the other hand, a plant species that has been found only in Shenshef. Most certainly, the discrepancy between these records is partly affected by the unbalanced excavations at both sites. The wild plants are represented by some 110 wild plant taxa and could only partly be identified to the level of species. Some of the botanical remains are still unidentified, and it is assumed that they concern both cultivated and wild plant species. A part of these still-unidentified remains have clear diagnostic features and may even concern whole seeds or fruits. Their identification is hampered by the incompleteness of the reference collection.
The cultivated plants were obtained from an extended geographical area, and for this reason a large number of economic plants from remote areas have to be taken into consideration. As far as the reference collection of the wild plants is concerned, the incompleteness is caused by the difficulties met in collecting the seeds and fruits of desert plants, their presence largely dependent on the fluctuations in the humidity. The first step to a successful identification of the unidentified cultivated plants would be the compilation of lists of potential plant species that were traded from specific areas. Such a valuable checklist was presented by Warmington , who selected medicinal plant products traded by the Arabs on the assumption that they were perhaps also known to the earlier Romans. Although the number of species is limited, this list includes the coconut , the emblic , and the tamarind , three species evidenced at Berenike by the current archaeobotanical research as real trade products. Although these three species were already present among the botanical remains unearthed in Berenike during the first excavation seasons, the identification of emblic was only successful several years later, when a sample of this food plant containing some of the characteristic endocarp fragments was obtained from a spice shop in Cairo. Consultation of Warmington’s list at an earlier stage would have been of great help. Recently, a more extensive list of plant species that might have been traded from tropical Asia and China has been compiled. The matter is urgent as spice shops in the Near Eastern bazaars are becoming increasingly threatened. Their broad and exotic selection is being replaced more and more by a limited and more standardized supply as a result of globalization. But it is still possible to get a glimpse of the range of goods offered for sale during the past and to collect relevant reference material. From the uneven distribution of many plants over the trenches as well as the still untouched large habitation and industrial areas of both settlements, we can expect more plant species to be unearthed in future excavations, which in turn may shed new light on the subsistence and trade economy. This is not to say that a more elaborate sampling will ultimately result in a representative picture of subsistence and trade. Despite the excellent preservation conditions, archaeobotanical research is faced with the transfer and consumption of plants that have left no visible marks and of which all traces are vanished. Wastage of expensive commodities would have been the exception rather than the rule.
Food may also have been prepared in such a way that allowed for no inedible leftovers. Other food is completely digested, and therefore no identifiable remains would have been excreted in the feces. Despite their botanical richness, a similar warning is called in for the analysis of dump areas. Most of the botanical remains that have been investigated originate from organic dumps, which appeared to be predominantly middens and fl at dump areas in the proximity of stone buildings. It should be realized that in the course of time even such concentrated and well-preserved archives become only a faint reflection of the organic waste once deposited. Several processes are responsible for the reduction of this organic archive. Although it is conceivable that at least part of the habitation area was inaccessible to straying animals, as was the case, for example, with locally constructed garden plots, it seems most likely that organic dumps in particular were frequently visited by browsing animals, as is evidenced by the many dung pellets especially of sheep and goats and, to a lesser extent, from camels. Archaeozoological research and recent observations of trash deposits have attested to the presence of the following animals: cattle, pigs, chickens, sheep, goats, donkeys, black rats, dogs, Egyptian vultures, vertical grow rack system and brown-necked ravens . On a smaller scale, small creatures, such as beetles and larvae, also eat organic material. This kind of reduction may even last for thousands of years, as could be observed when special attention was paid to this process while analyzing a Roman trash deposit in Karanis . It was found that such small animals are present at least in the upper 10 cm of the trash. The reduction of the archaeobotanical archive may also be the result of burning the trash. During the excavations campaigns, such burning is still frequently practiced to get rid of the stench and to control as far as possible the number of flies, rats, and other such vermin, which in turn attract predators such as snakes. For safety reasons, such burning would mainly have taken place outside the settlement. This explains why most of the plant remains from the trash dumps inside the settlement have been preserved by desiccation, those used for offering ceremonies being an exception. The wind would also have been a factor in the reduction of the trash deposits. When trash dries out during high temperatures, it may be easily blown away. Only those remains that were sufficiently covered would have been protected from further losses due to wind and other factors, and these deposits constitute the actual archive that has been sampled . Some prudence is called for when numbers of plant remains are compared. A quantitative approach is not only obscured by the different plant parts we are dealing with, which is even true for seeds and fruits, but also because numbers of seeds and fruits differ highly among plants. Additionally, taphonomic processes bias the number of sub-fossil remains. Some fruits may serve as examples. It appeared that the endocarp fragments of the almond easily fall apart in an arid environment. This is caused by the relatively thick veins that are present in the middle of the inner fruit layer. Differences in the fruit anatomy also cause a discrepancy in counts. Each consumed date and olive , for example, will produce only a single leftover, either a seed or a fruit stone . Other fruits contain many seeds, such as a pomegranate with has some 250 to 400 seeds and the fig , whose number of “seeds” ranges from about 700 to 1,800 . A simple conversion of the number of sub-fossil seeds that have been found to the original number of fruits that were consumed is therefore impossible. The comparison of the retrieved plant remains is further complicated by differences in the consumption of fruits. Seeds of the date and fruit stones of the olive are not consumed and have a good chance of ending up in the trash deposits. On the other hand, the small seeds of the pomegranate and the fig are consumed and excreted and are therefore less likely to be found in trash layers. The consumption of whole, destoned fruits at the sites proper would have been the main source of recovered fruit-stone fragments. The concentration of these remains in middens or dump areas will partly depend on the way the fruits are eaten. Eating olives with an Egyptian colleague in a courtyard, I observed that I just threw the fruit stones on the ground, while my colleague collected all the fruit stones carefully but simply discarded the packing material of his cigarettes. From an archaeobotanical point of view, a possible recovery of my discarded fruit stones during a hypothetical excavation is most unlikely, whereas in excavating the trash deposit, one would have a good chance of finding some of my companion’s, which most probably will have been deposited there by the kitchen help.