'Small Fields or Big Fields?' That is the Question

They have not yet been crystallized but depend on where future evidence will lead us. Our thesis here is that it might be possible to ascribe 'fields' to one particular crop plant of the West House (Fig. 5). More precisely, when we say that we can ascribe 'fields', we mean that we will try to make hypotheses on the possible size of fields, whether the fields were small or large. Our basic assumption is that the crop of one field is reaped, threshed and stored together.

For this investigation we used a particular crop plant, Lathyrus clymenum L. (Fig. 1), commonly known as Spanish vetchling, which is found stored in the West House. There are, in all, seven samples of this crop in this house, and it seems to have been one of the most popular crops in this building.

Akrotiri is a unique Late Bronze Age site. For those who study macrofossil plant remains, the reasons for its importance are that it offers a very high potential for (a) clarifying certain taphonomic issues in archaeobotany, due to the unique macrofossil preservation; (b) the use of archaeobotanical data to gain understanding of the use of space and the use of objects; (c) the use of archaeobotanical data to gain insights into Bronze Age society, economy and agriculture.

This last point, and in particular one aspect of it, the discussion of whether stored crops such as Lathyrus clymenum in the West House reflect small or big fields, is the subject of this paper.

THE TAPHONOMIC ISSUES

The Late Cycladic I destruction horizon, and in particular the destruction of the West House at one point in time, is indisputable. The archaeological deposits immediately underlying the ash-fall are therefore contemporary, and for those who study plant remains this implies that all stored crops are also contemporary, which means that all stored crops are the products of one or, at most, two years of cropping. We know that crops such as cereals and pulses could not be stored for longer periods of time without excessive loss of seed due to weevil attack and other infestation.

The second point is that the catastrophe occurred throughout the entire settlement simultaneously, and supposed temperatures of c. 250°-300° C touched the site (Tarling, pers. comm.). Since no evidence of fire has been detected in the West House, we can claim that all organic matter was exposed to approximately the same range of temperatures with only minor micro-ecological variations. This is an important factor as it is well known that the preservation and the size of macrofossil plant remains is affected by their exposure to temperature and duration of charring as well as by the moisture content of the seeds at the time of the charring process. The fourth variable, whether or not seeds were charred in reduced oxygen, is not detectable on charred seed remains, and, therefore, will not be considered here. Here we shall try to discuss each of the three variables which could affect the size of seeds.

TEMPERATURE

It has already been mentioned above that the temperature of the tephra which reached the West House and, indeed, the entire settlement of Akrotiri, would have been in the range of c. 300^o C, but due to localized conditions such as humidity in the rooms, protection of seeds by storage jars, and possible evaporation of the heat, seeds would perhaps have been exposed to slightly lower temperatures. How much lower is difficult to say. We can only say that the constituents of wood, hemicellulose, cellulose and lignin, are reported to have individual breakdown temperatures, and in the case of cellulose the range reported is between 240^o and 350^o C (Jenkinson 1976, 3). It is probably indicative that in the West House there was no preservation of wood and therefore we could assume that the temperature might have reached the fringe of this range. The other restricting factor is that 250^o C seems to be the lowest temperature which could char seeds. Bowman (1966, 14) experimented on grains of Triticum aestivum and has shown that the lowest temperature which could char seeds was 250^o C. Below that temperature seeds were not charred, irrespective of the duration of heating, whereas at 250^o C the difference between charred and uncharred was affected by the duration of exposure to heat. His experiments show that times up to half an hour at 250^o C did not char seeds and that charring occurred after 1 hour. On the basis, therefore, of the lack of charred wood and the presence of charred seeds, we could assume that the West House was exposed to a temperature of approximately 250^o C and probably less than 300^o C. The uniformity of the uncharred glume parts of the cereal grains in all samples of the West House also indicates a temperature not higher than 250^o C (Bowman 1966, 10). However, it would be interesting to submit some of the seeds to Electron Spin Resonance (ESR) Spectroscopy to determine the thermal history of the Akrotiri seeds (cf. Hillman et al. 1985).

DURATION OF HEAT

The surface lustre of the seeds indicates periods of exposure to heat; for example, the dull lustre of seeds has been shown experimentally to indicate longer periods of exposure to heat than seeds with a shiny lustre (Jenkinson 1976, 44). Jenkinson stated that lentil seeds heated for 10 hours had a duller lustre than seeds heated for a shorter period. In the West House, the Lathyrus clymenum L. seeds could be described as having a dull lustre and as they were stored in storage jars, most probably would not have been exposed to direct heat and less evaporation, since the jar would have behaved as a small oven. A point worth remembering from what has been mentioned above, is that at 250^o C seeds do not char in less than 1 hour, therefore, we can assume that exposure to heat was for at least 1 hour before its evaporation.

MOISTURE

Both Jenkinson (1976, 46) and Bowman (1966, 15) claim that a high/low moisture and 6-10 hours heating is indistinguishable in dimensional terms. Therefore, moisture content is not a factor to be considered when comparing measurements of the seeds found in the West House. In an environment such as the Cyclades, one would believe that the moisture captured by the stored crops would not have been very high and, moreover, of equal value. Measurements taken over 10 years at Butzer Farm in the south of England (Reynolds 1979, 72) demonstrate that grain is harvested at an average of 16% moisture content. This percentage would have been much lower for Akrotiri, but might have reached these figures in damp basements after months of storage in the house. It is assumed, however, that the difference in moisture level between samples was not great enough to produce visible effects on the size of seeds of the West House.

We could, perhaps, sum up and say that all indications point to temperatures in the range of 250^o C, if we base our assumptions on the estimations of temperatures by Dr Tarling, the absence of charred wood, the presence of charred seeds and the presence of uncharred glume parts (awns, rachis, lemmas) in the same samples. The duration might have been in the range of one hour, and we can say at this stage that seeds from every room of the West House had probably been exposed to a uniform charring process and are comparable.

LATHYRUS CLYMENUM L: THE PLANT AND THE SEED

Lathyrus clymenum L. is a plant that still grows today on Santorini and is considered a staple food by the inhabitants. The crop is locally named arakas and when it is split to make their staple dish is renamed fava. Perhaps the term fava misled Friedrich (1983, 243) in misidentifying this crop as 'fava beans' (see Friedrich et al. 1990), i.e. Vicia faba ssp. minor, otherwise called 'Celtic beans'.

Scanning Electron Microscope (SEM) work on both archaeological and modern specimens of Lathyrus clymenum is, unfortunately, not yet conclusive. Although, initially, it was believed that a distinctive surface pattern of the testa (Fig. 2 and 3) could be identified as regular, blunt papillae were detected on both archaeological and modern seeds. Ann Butler (Ph.D. in progress, University of London), who is presently studying the surface sculpture of Viciaceae, claimed that our results could not be considered conclusive as these features were not yet characteristic to species level, but could include other Viciaceae.

Morphologically, the seed of Lathyrus clymenum is very different from Vicia faba L. The hilum of the latter is on the shorter side of the seed, whereas in Lathyrus clymenum it is on the long side (Fig. 7).

Davis (1970, 361) stated that plants of Lathyrus clymenum and Lathyrus articulatus L. could not be separated as different species in the Mediterranean. On the other hand, Tutin et al. (1968, vol. 2, 142) claim that the pods of these two species are different. The pod of Lathyrus clymenum L. is described as 'channelled on the dorsal suture' and 'not torulose', whereas Lathyrus articulatus L. has a pod which is 'not channelled on the dorsal suture' and 'somewhat torulose'. The legume which is grown today on Santorini complies with the first description, i.e. Lathyrus clymenum L.

At present, therefore, the morphological criteria of both modern and archaeological seeds, as well as the botanical identification of the modern, cultivated plant of Lathyrus clymenum L. are the only criteria used for identification purposes here.

The shape of the seed of Lathyrus clymenum L. could be described as oblong with three rounded edges, except for the edge on which the hilum begins which is situated on the long side of the seed. This last edge is mainly angular. This excludes all archaeological examples of legumes identified and published to date in the Mediterranean, and excludes amongst others Vicia faba ssp. minor, as well as Lathyrus sativus L., and Vicia sativa (grass pea) (Fig. 7).

LATHYRUS CLYMENUM L: CROPS

As mentioned above, there are seven samples of this crop found stored in the West House (Fig. 5), all of which come from Room 5 (ground floor), with the exception of sample 65 which was stored in Room 3C. This number does not include the imprints believed to have also been of this crop from Area 7 (first floor) (Fig. 6). Sample 65, however, is somewhat different from all other samples in that it includes more split legumes by weight than whole ones, and the unsplit seeds of this sample are the smallest seeds of all samples. This probably implies that the sample of Lathyrus clymenum L. had been split, just as they do on Santorini today before the preparation of their staple dish, fava.

Since all these crops were found in pots (Fig. 5), the seeds are considered to have been equally protected, and none were water-sieved. They were only dry-sieved with sieves of mesh sizes of 1 mm and 300 μm. We also aimed to study at least 50 grams of each sample. Where there was less quantity, we examined the total quantity present.

Our aim was to see whether samples of Lathyrus clymenum L. taken from different pots (different samples) indicated in any way that these also came from different fields. The variables used as indicators of different fields are the following: (a) size of the seeds of Lathyrus clymenum L.; (b) species of contaminants; (c) size of major contaminants; (d) species (range) of weed seeds; (e) weed infestation level.

     (a) Size of the seed of Lathyrus clymenum L.:      Analyses of variance were done for the length, breadth, thickness, ratio of L/B (length to breadth) and L/T (length to thickness) (Table 1). No marked differences were noted between samples, yet it was evident that sample 20/29 had small Lathyrus clymenum L. seeds and sample 65 had very small seeds -the smallest of all samples. Therefore, the size of the crop, except for these two samples, was not helpful at this stage in dividing our samples into groups.

     (b) Species of contaminants within these crops:      The major contaminants in this crop (Lathyrus clymenum L.) were Lathyrus cicera/sativus. They were the most abundant in all samples. Nevertheless, they appear in different percentages of contamination (Table 2). However, samples 5/10, 16 and 20/29 have smaller percentages (12.7%, 11.0% and 12.0% respectively). When it comes to considering the second and third most abundant contaminants, the first sample, 5/10 differs, but samples 16 and 20/29 reflect great similarity in that lentils are the second most abundant contaminants with percentages of contamination of the range of 2.6% and 2.7% respectively. The third major weed in these samples is Sherardia arvensis, with a contamination level of 2.0% and 1.5% respectively. A further factor considered is the general percentage of contamination by weeds for these two samples (16 and 20/29), which is 3.7% and 3.2% respectively. We can say, therefore, with some degree of assurance, that these two samples could be grouped together, despite the difference in the size of the crop plant (Lathyrus clymenum L.). Other samples which could be comparable are samples 9 and 14. In these the size of the crop plant, Lathyrus clymenum L., does not differ and the degree of contamination for Lathyrus cicera/sativus is comparable (2.0% and 2.1% respectively).

Therefore, out of the seven samples of Lathyrus clymenum L., we have been able to isolate, on the basis of contaminants, four samples which could form groups, samples 16 and 20/29 (group A) and 9 and 14 (group B). These could have come from the same fields. Sample 1, on the basis of contaminants, stands very much on its own. The percentage of contamination by Lathyrus cicera/sativus is the highest of all samples, 16.5% (Table 2), and contamination by the second and third most abundant contaminants does not resemble any other sample (Sherardia arvensis 6.3% and Lens 4.2%). The other two samples which could have been grouped together are samples 5/10 and 65, but at present we lack more evidence. They have similar contaminants, such as Lathyrus cicera/sativus (12.7% and 11.6% respectively), as well as comparable weed contamination (2.8% and 2.9%) but unfortunately sample 65 is still too small to provide strong evidence in that direction.

     (c) Size of major contaminants:      Although it is believed to be a variable worth considering in the future, we presently do not have enough evidence to use this variable for the sake of comparison. More samples need to be measured.

     (d) Species (range) of weed seeds:      This variable reinforces the association of sample 16 and 20/29 (Table 4) (group A) whereby the number of species is comparable (11 and 10 species respectively). It does not, however, reinforce the association of samples 9 and 14 (group B), though counts need to be made for these two samples. The other sample which stands out, as in all other variables, is sample 1 where the number of species is 21 (Table 2).

     (e) Weed infestation level:      The weed infestation level also reinforces the association of samples 16 and 20/29 (group A). Samples 9 and 14 (group B) are also comparable (0.5% and 0.9% respectively). Even the percentage of contamination of the crop is somewhat comparable (5.3% and 4.5%). Sample 1, as usual, stands out as being very different, with a contamination of 7.4%, which is the highest of all samples. However, even though the contamination by weeds of samples 5/10 and 65 (group D?) is very similar (2.8% and 2.9% respectively), we still need to process more of sample 65 to verify this. Moreover, the number of species of the weeds (13 and 14 respectively) is very dissimilar.

Therefore, on the basis of the above-mentioned variables, the Lathyrus clymenum samples could be clustered into 3-4 groups (Table 4).

FIELD SIZE: CAN WE MAKE ANY HYPOTHESIS?

Present day data from Santorini for the quantities of seed produced are between 300-600 kilograms per zevgaria (one zevgaria represents three stremmata, which is equal to 3,000 m²). The lowest number represents a bad yield and the higher a good one. As pulses are heavier per unit than cereals, all calculations were based on volume.

For the sake of this present exercise, since complete accuracy is neither demanded nor possible, we shall accept that 1 kilogram = 1 litre, as we have calculated the volumes of pots in litres. Table 3 indicates the sample numbers, the equivalent pot type and volume. The groups that were formed using the above-mentioned five variables (a to e) indicate that no group was stored in pots which had a greater volume capacity than 120 litres which, probably, indicates - if we accept that yields were equivalent to today's on Santorini - that these crops/samples do not come from one large field but rather from fields whose crop production lies in the range of 100-130 kilograms per yield. This represents fields of an approximate size of 1,000 m², which is equivalent to today's stremma in Greece. In other words, it does indicate small fields.

CONCLUSIONS

More calculations are presently being conducted on samples from the West House which should either validate or not all groupings formed, as well as the statement about 'small fields'. Moreover, the indication of 'small fields' for the West House does not, automatically, indicate 'small fields' in other houses of the same settlement. The same type of work needs to be conducted in different houses, so that the general pattern of farm holding would emerge.

What the full meaning of 'small fields' is in terms of economic and social organization still awaits explanation. In terms of economic organization, 'small fields' could be a safe way for farmers to minimize risk of crop failure, since it could indicate varied crop production. Small and scattered fields would encourage the cultivation of varied crops depending on soil and other conditions. Secondly, small fields belonging to one 'household unit' would control different micro-ecological niches, which would, possibly, encourage this varied crop production, or even, if it is the same crop which is being cultivated, produce differently, due to a multiplicity of environmental factors, primarily soil and micro-climatic differences. Thirdly, small fields would provide a wider availability of good land and a more equable distribution too. Thus, fragmentation of fields can have a sound economic base, especially in an age when the factor of time was perhaps not treated as a serious variable affecting decision making.

However, we are still a long way away from resolving these matters, and here is not the place to discuss the social implication of 'small fields' for the West House. Questions rather than answers spring to mind. Did the owners of the West House own fields from which they received the Lathyrus clymenum crop? And did they tend them with the help of 'slaves'? Or was the land fragmented because it belonged to independent small farmers who traded their produce with Akrotiri in exchange for other produce? Here are 'small fields' for thought.

Addendum

In the discussion that followed the delivery of the paper, Dr S. Limbrey suggested that what were referred to as charred seeds were possibly humified seeds. Pending ESR work which is intended for the Akrotiri seeds and particularly for this group of seeds referred to in the paper, we can only rely on descriptive characteristics in order to refute humification of these seeds. These seeds are black (not dark brown), very brittle and produce a typical black dust, the type one gets from charcoal. The description given in a private conversation with Dr S. Limbrey led her to agree that these seemed more probably charred.

Professor D.H. Tarling, in a personal discussion, indicated that he has reconsidered the temperature of the tephra which touched the site of Akrotiri; he believes now that the temperature must have been lower, in the region of approximately 150^o C which seems to clash with the presence of charred seeds at Akrotiri. It would be very interesting to see what the ESR work will estimate.

----------------------------------------------------

---------------------------------------------------