Oilseed Cakes in Italy and France: Opportunities and Difficulties of a Market (late 19<superscript>th</superscript> and first half of the 20<superscript>th</superscript> Century).

This paper addresses the trade and commercialisation of oilseed cakes (residues from the extraction of oils) and press cakes in Italy and France during the last decades of the 19^th century and in the first half of the 20^th century. It tries to demonstrate that the diffusion of oilseed cakes for livestock, a distinctive sign of the intensification of breeding that involved all of Europe, or as organic fertilisers, took place at the crossroads of multiple dynamics. Trade policy of the states, industrial choices and development paths of the different rural worlds help to explain the variations in timing, spatial scale and methods used. The spread of oilseed cakes confirms that the modernisation of European agriculture happened on different and interrelated fronts.

Keywords: rural history; Italy; France; oilseed cakes; modernisation; countryside; livestock; feeding; fertilisers; Agrargeschichte; Italien; Frankreich; Ölkuchen; Modernisierung; Landwirtschaft; Viehbestand; Fütterung; Düngemittel; N 53; N 54

This paper addresses the trade and commercialisation of oilseed cakes (residues from the extraction of oils) and press cakes in Italy and France during the last decades of the 19th century and in the first half of the 20th century. Processing oilseed was part of the larger industry of oil and fat materials, which spread broadly, from lighting to industrial uses, for food as well as non-food purposes.

Oilseed cakes were used as livestock feed or as fertilisers in farming (in particular those made from exotic rapeseed, mowra, poppy, castor seed etc.). Although there were overlaps and close relationships in the production circuits of these two products, it is necessary to specify, in fact, that not all the oilseed cakes were edible. This article is devoted, in particular, to oilseed cakes usable as food for livestock, in a period in which in many European countries and in the US, new patterns of modernisation emerged: a central role was played by livestock, considered to be a central element in the economic development of the countryside from 1880 onwards, and even more so in the interwar period. The study of the history of oilseed cakes therefore lies at the crossroads of multiple research paths: first, the study of agronomic techniques for improving and increasing livestock (number and weight); secondly, the study of the diffusion of cakes as a fertiliser in fields; in the same way, the production processes, the supply circuits of the necessary raw materials (oilseeds) deserve to be examined in depth, as well as thirdly the marketing circuits of oilseed cakes. In particular, the great variety of the production of cakes and their different purposes (fertilisers, fodder) triggered multiple marketing paths, which are not easy to reconstruct. Each of these three aspects intersects with some fundamental questions concerning the tools to improve agriculture in the 19th and 20th centuries, such as quality control and fraud suppression, chemical fertilisers as opposed to organic and customs tariffs.

This paper presents observations on the roles of the elements involved in the production, trade, and sale of oilseed cakes. In the second section I analyse the nature, use and importance of seedcakes in the Italian and French rural worlds. In section three, starting from the observation that the seedcake market was tied to the development of oilseed production and trade, as well as to the dynamics of international trade, I will attempt a brief outline of the seedcake trade in France and Italy, seeing that France, at the time, did not absorb the full production of seedcakes. In section IV, I shall work on a micro scale, analysing the situation in Marseilles, so as to identify the interaction of the interests of the farmers and the other stakeholders involved in the oilseed cake trade. Keeping in mind the underlying issues of contrasts between different interests in relation to fodder by producers, farmers, traders, and the spread of synthetic fertilisers, the conclusions focus on the ever more pressing difficulties that hindered the circulation of seedcakes.

The comparative study of French and Italian cases makes it possible to highlight two different paths of modernisation. In these different paths, both the exogenous factors in the rural world (the states' commercial policies, the ambitions of the imperialist countries) and the endogenous factors played a decisive role. Among the latter, it is necessary to underline the importance of three fundamental issues: the knowledge produced by the agronomic sciences (which oilseed cakes? Of what quality? How to convince and reassure peasants and farmers about their purchase and use?), and the relationships between these and the structures of modernisation (e.g. French Syndicats) ; the specific conditions of the production industry and the relationship between this and the national market; the rural contexts to which cakes were destined.

Despite the fact that France, compared to Italy, had a more rooted tradition in the use of cake, the times and modes of cake production proved to be significant constraints to its full use in the countryside. The production of cake was closely linked to the marketing and location of seed processing, which had an impact on the type and quality on offer. Although with a different rural background, even in Italy the events linked to the production of cake-pieces played a significant role. The presence of a different industrial apparatus connected to the processing of seeds and the processing of cake made these products less accessible to Italian farmers. This happened also because of a lesser knowledge about the use of these products in agriculture, both concerning the feeding of cattle and the fertilization of the fields, even if there were some innovation niches in various parts of Italy. In the Italian case, however, the role of the agricultural context and, more generally, of the rural context had the greatest impact. The distrust towards the use of oilseeds, in fact, had ancient roots and was inseparable from the historical preference accorded for the cultivation of the olive tree and its derivatives. In fact, every innovation would have had to deal with the rural roots and with the economic interests linked to the olive and its oil.

Oilseed cakes (tourteaux in French, as well as tourtes or pains d'huile, panelli, or panelle, or panelli di semi oleosi in Italian) are part of the category of concentrated feed used both as nourishment for animals and to fertilize the soil. They are produced by compressing the solid residues left from the pressing of seeds or fruits, either in their natural form or with their coverings removed, to extract oils that can be used as fodder, in industrial processes or pharmaceutical products. Given the similar forms in which they are marketed, the same term is also used for cakes made from cereal flours, such as corn. Used in the Flemish region of Hainaut, as early as the start of the 18th century, the seedcakes started to be used in the French countryside in the mid-19th century, but did not become a part of Italian agronomic discourse until relatively later. It was not until the end of the 19th century that their use spread to the point of gaining notoriety on both sides of the French-Italian Alps, soliciting chemical analysis.

In 1876, in an official report by the Italian Ministry of Agriculture, which collected the reports and memoirs produced by different local institutions (e.g. academies, comizi, professors), the Comizio of Parma, wrote rather despondently: "cinders and green manure are little used, the oilseed cakes are given to the oxen that are fattened, instead of to the lands as a fertiliser". Using oilseed cakes as a cattle feed was sometimes regarded as a diverging use from their primary purpose, which would have been to fertilize the land. In a 1901 study published by Giuseppe Fascetti, the use of oilseed cake in the feeding of livestock was thus defined as a necessary choice for lack of a better solution, or in the absence of the traditional systems of feeding, or when hay was too expensive:

"This happens when haymaking in farms is anomalous and fodder fails and requires that the nutritional value be raised; when it is necessary to correct some industrial residues of a large nutritional value that are to be used as feed; or when the stocks of hay are very diminished until the end of the season, are not enough to last until the new harvest and the farmer is led to limit its consumption. This is the most general and common case because it occurs almost with a certain constancy at the end of the winter of each year and it occurs more frequently when an irregular season causes a late forage production. Then the hay prices rise in an exceptional manner, at 12, 13, 14 lire per quintal. These prices disturb the farmer, who is forced to restore his limited reserves with new provisions and cannot use hay without detriment. In these circumstances the concentrated foodstuffs in general and oilseeds cakes in particular represent for him the way to salvation. But the difficulties that arise for the farmer in preparing the rations from an economic point of view, the nonjudicious choice of corrective foods, their economic value that does not always correspond to nutritional value often determine opposite effects to those that are in the economy and in production".

In the same report of 1876 cited above, with reference to the Venetian situation, it was written as follows: "Fertiliser factories do not exist, only the oilseed cake produced by castor oil factories, which is largely exported and now [the farmers are beginning] to employ a few hundred quintals to manure lands destined for the cultivation of hemp". The use of oilseed cakes as fertilisers for the land devoted to hemp is also confirmed for the countryside of the provinces of Bologna and Modena, where this cultivation had more ancient roots.

Except for some well localized territories, Italy seems to have been in a different position compared to what was happening in other European countries. Italy participated in the global process of transformation of the feeding of cattle with an original position: as a small consumer and a small exporter at the same time (Fig. 1 and 2). The figure is even more marked if we consider that, contrary to what happened to other materials and products useful for agriculture (fertilisers, copper sulphate, machinery), in which there was a chronic deficit of the trade balance, in the case of oilseed cakes, the value of exports exceeded that of imports, albeit within a more modest business.

Graph: Fig. 1 Import-Export of Cakes from Walnut and Other Materials in Italy, 1870-1913 in Metric Tons. Source: Bankit-FTV, database of Italian foreign trade, 1862-1939, collected by Bank of Italy under the scientific direction of G. Federico, G. Tattara, M. Vasta, see G. Federico/S. Natoli/G. Tattara/M. Vasta, Il Commercio Estero Italiano 1862-1950, Collana Storica della Banca d'Italia; Banca dati Bankit-FTV, Roma-Bari 2011, p. VII), https://www.bancaditalia.it/pubblicazioni/collana-storica/statistiche/statistiche-04/CSBI-statistiche-04.pdf, 22.08.2020.

Graph: Fig. 2 Export of Cakes from Italy to some other European Countries, 1870-1913 in Metric Tons. Source: See Fig. 1.

The articles, pamphlets and texts dealing with the production, marketing and consumption of oilseed cakes, during the 1880s, tried to promote the use and the qualities of this product, in particular, as feed for livestock, while less attention was paid to those used as fertilisers. A significant example of this model, which highlights the important role played by the scientific milieu in the dissemination and modernisation of agriculture, is the study by Ferdinando Rossi on cotton cake for feeding cattle.

Rossi started from the initial consideration that the cultivation of cotton in Italy was limited only to 14 southern provinces and the potential production of cakes was still rather limited. After recalling the larger diffusion of this product in England and the United States, Rossi went on to comment upon the results of the chemical analysis of oilseed cakes: he evaluated the comparative cost of the individual nutrient units in the different types of food for livestock, and then moved on to investigate the outcomes in terms of weight and production of milk, reporting on the experiments carried out by Prof. Giovanni Celli at the Regio deposito di animali miglioratori of Portici (Naples) and in France by Professor Villepin, published in the Journal d'agriculture pratique (1872). These three points are present almost in all publications of this type.

The examples could be multiplied. The encouragement of the use of oilseed cakes always took place within the battle for rational agriculture, which found ample space in the publications of the time. Reporting on Villepin's research, Ferdinando Rossi recalled that the cost of the nitrogen needed for one kg of beef was 0.108 lire in the case of the use of cotton seed cake, compared to the 0.672 lire for barley. Giuseppe Fascetti reported the results of his experiments in the stable, expressed thanks to the collaboration of a farmer of Lodivecchio (Lodi), demonstrating that the introduction of linseed cakes at moderate doses actually saved the cost of the ration for animals, while maintaining the level of milk production, and that corn cakes, more pleasant to the taste, stimulated a greater milk production, compared to a cost aligned to that of traditional hay (13 lire). In both cases, therefore, an advantage was demonstrated in the use of oilseed cakes, albeit in different forms and ways. There was no lack of warnings and cautions, however: the oilseed cakes had their most evident effects in the increase in milk production in the first period of use, while they lost effectiveness over time. For this reason they were ideal for supplementing daily rations for a limited period, perhaps coinciding with a change in the usual diet, in order to maintain or supplement milk production levels. Antonio Cugnini showed that a particular type of sugary seedcake (panello zuccherino, the so-called type B produced by the Italian distillery company: peel of grapes flour, dry borlanda and molasses – which were by-products of beetroot), could usefully replace that of maize, because they were both cheaper and more pleasing to the taste of cattle. Chemical analysis allowed for the precise calculation of the nutritional units for each type of oilseed cake. A comparison with the market prices of the other seedcakes allowed the calculation of the cost of the individual nutritional units for each type. In this way it was possible for Cugnini to prove that the single nutritive unit of the sugary seedcake was the most convenient on the local market, namely that of Reggio Emilia. The nutritional unit of the sugary seedcake in fact amounted to 0.137 lire, compared to the 0.182 lire for flax cake, and the 0.156 lire of the corn cake, or for wheat bran (0.189 lire).

Starting from the end of the 19th century, more and more organic systematizations of the material appeared, in particular with the work of Antonio Sansone, which collected and integrated work published previously in the review L'Italia agricola. Giornale di Agricoltura, and later, with the text of F. Tornar and G. Giacomelli. A few years earlier, (in 1925) a general law was passed to deal with fraud in the preparation and trade of agricultural products, which in Art. 9 dealt specifically with the case of cake preparation.

In France the production and consumption of oilseed cake had long assumed greater importance than in Italy. In this sector, France could draw on the preferential supply channel of its colonial territories, as well as a solid oilseed and chemical industry, with Marseilles undoubtedly standing as one of the cornerstones of this trade. The supplies arriving from Africa largely made up for the lower domestic production of oilseeds.

As shown by the studies on the industrial set-up of the city of Marseilles, although the oilseed cakes were a by-product of the oil industry, they repeatedly played an important role. The logics of exchange that dominated were those necessary for the proper functioning of the supply of raw materials in Africa. In the period of the great increase of the oilseed trade with the west coast of Africa, the Marseilles traders exported most of the oilseed cakes produced in Marseilles to Liverpool (from the 1840s to the mid-1870s), in exchange for textiles, needed in turn to sustain the trade with Africans, to which the textiles were sold.

The share of oilseed cakes exported to England decreased significantly after 1877, when consumption in the countryside around Marseilles reached ever more consistent levels. This confirmed a trend already present for some decades, as demonstrated by the recent study by Laurent Herment and Éric Mermet.

The surrounding countryside, however, could not absorb all the production, which continued to be sold in northern Europe in particular, even when the global economic situation changed the routes and the actors in the fats trade. Beginning in 1873, a significant crisis emerged in the African peanut trade due to the significant growth in peanuts from India reaching the European market. The conditions under which the development of the oil industry and oilseed cake manufacturing took place in Marseilles were subject to profound change and involved, alongside other factors, a long period of crisis from 1882 to 1897.

Even in this new situation, the role of the cakes was of great importance. As explained in the seminal work of Daumalin, it was precisely the search for palm kernel oil, which was used in large quantities for feeding cattle in Germany, that determined a turning point in the market for fatty matter. This growing demand in Northern Europe, which was present also during the 1920s and the 1930s, determined the great development of Hamburg as the main port for the importation of palm kernels. As a whole, the production of oilseed cakes in Marseilles constantly grew throughout the second half of the 19th century.

The oilseed cakes used on the domestic market included both lower and higher quality. Among the latter, those made from peanuts were considered to be among the most highly valued, and were also largely requested by foreign buyers. During the decade that preceded the crisis of the peanut trade with the African coast mentioned above, the domination of peanut oilseed cakes was eroded by the increasing trade in sesame cake, both of copra, palm kernel, rapeseed and turnip rapeseed. The case of Marseilles shows how the production of oilseed cake was affected by the dynamics of seed imports (Table 1).

Tab. 1 Importations of Oilseed and Production of Oilseed Cakes in Marseilles, 1868-1879 in Metric Tons.

1 Source: Own calculations, based on Chambre de Commerce de Marseille , Compte-rendu de la situation commerciale et industrielle de l'arrondissement de Marseille, Various years (in brackets the year of publication): 1868 (1869), pp. 52-53; 1869 (1870), pp. 60-61; 1870 (1871), pp. 58-59; 1871 (1873), pp. 48-49; 1872 (1873), pp. 44-45; 1873 (1874), pp. 56-58; 1874 (1875), pp. 64-65; 1875 (1876), pp. 70-71; 1876 (1877), pp. 61-62; 1877 (1878), pp. 65-66; 1878 (1879), pp. 57-58; 1879 (1880), pp. 52-53; Daumalin , Marseille et l'Ouest africain, pp. 85, table 9, 355-358, table II. * OTH: other oilseed (included copra, palm kernel, poppy, rapeseed and turnip rapeseed).

Even in France, in the first quarter of the 20th century, there were some attempts at systematically studying the composition of oilseed cakes on the market. The works of Léon Bussard and George Fron (1905), and Léon Bussard and Charles Brioux (1925), allow us to have a complete picture of the types and qualities of the cakes available on the French market. Table 2 summarises the main types of seedcakes available on the French market in the years between the two World Wars. The system of classification used is that proposed by Bussard and Brioux, though the format, as the authors themselves concede, is somewhat contrived, seeing that to a large extent the native species were actually imported from abroad.

Tab. 2 Presence of Seedcakes in France, 1925.

2 Source: Bussard/Brioux , Tourteaux, pp. 4-6.

The increasing attention paid by government authorities to oilseed cakes in the period between the two World Wars may be attributed to different reasons: the noteworthy quantities of nutrients they provided, resulting in their widespread use in farming; the extensive variety of seedcakes in circulation; the elevated level of French domestic production (649,428 tons in 1931; 637,216 in 1932); plus their noteworthy commercial value (sales prices abroad peaked in 1926 in Italy, at more than 101 lire of the period for a hundred kilos, while the top price in France was reached in 1929, at more than 122 Francs/one hundred kilos). The primary objective was to control two areas of dispute that often clashed in both public debate and the courts: rising prices and efforts to suppress fraud; these two issues were of general interest to the entire sector of fertilisers, both organic and chemical, in France and Italy, as well as that of fodder supplies. This is illustrated by the fact that the preface to Bussard and Brioux's book was written by Eugène Roux, the Director of the Services for Health, Science and the Suppression of Fraud of the French Ministry of Agriculture.

Certain types of oilseed cakes were especially prized. One example was the high-quality peanut of Rufisque (a city in modern-day Senegal) processed by the Huileries Maurel & H. Prom & Maurel frères, a firm headquartered in Bordeaux and Marseilles. The peanuts were imported to Europe in their shells, guaranteeing greater freshness and less risk of their turning rancid. This raw material was used to make two products: oils for human consumption and seedcakes for use as livestock feed, with the further possibility of grinding the seedcakes into flour. According to the firm Maurel & Co., the chemical properties of the peanuts from Senegal, together with the special care taken in transporting and storing them in France, made for a top-quality livestock feed. In particular, their elevated nitrogen content (over 50 percent, according to the analyses performed by the Agronomy Institute of Bordeaux) ensured that the quality of the peanuts from the western coast of Africa proved far higher than those from India (Bombay, Coromandel), the Congo or Mozambique. According to other analyses cited in a promotional brochure put out by Maurel & Co, the seedcakes made from Senegal peanuts were five times richer in terms of their protein and albumin content, in addition to having three times the fat content of ordinary meadow hay; in terms of nitrogen content, 100 kg of top quality Senegal peanuts provided the same levels as 382 kg of wheat bran. The figures presented by Maurel & Co. were confirmed by the results of analyses carried out by other agronomic institutes of the period as well (Table 3).

Tab. 3 Chemical Composition of Certain Types of Cakes in Percent.

3 Note : the figures listed are the averages of those reported by the source for a given product. When the agronomic station that carried out an analysis was specified, its identity is indicated (see Source). Source : Bussard/Brioux , Tourteaux, p. 142 (rapeseed, agronomic station of Rouen), 265 (peanuts, agronomic station of Seine-Inférieure), 268 (soy bean, agronomic station of Seine-Inférieure), 273 (Sesame seed of the Levant), 286 (Sunflower seed), 295 (wall nuts), 321 (corn, agronomic station of Rouen), 323 (rice, agronomic station of Rouen).

The analysis of the chemical components of oilseed cakes paid attention to different aspects, depending on the type and the utilisation of oilseed cake. The nitrogen matter was important for oilseed cakes used as fertiliser; for cakes dedicated to feeding livestock it was also the fat content (in addition to that of nitrogen, being an indication of protein content) that was considered an important element. These components resulted in different prices, depending on the types of oilseed cakes. A specialized newspaper like L'Engrais, in 1897, for example, published a table in which it compared the oilseed cakes available in the marketplace of Dunkerque, in the north of France. For edible cakes, the minimum and maximum values of protein and fat matter were indicated, followed by the prices of single oilseed cakes; for cakes used as fertiliser, on the other hand, the parameters taken into consideration were the percentages of nitrogen and phosphoric acid. The prices of edible cakes were on average higher than those of fertilisers.

Having illustrated the main features of seedcakes, this section shall take a closer look at how they were traded. Prices on the French market for seedcakes did not, in fact, accurately reflect the outstanding nutritional levels. Soy cakes, for example, practically disappeared following the First World War (whereas they continued to be widely used in America), in the same way as the use of sunflower cakes was largely limited to Central and Eastern Europe. In contrast, cakes of copra (the hollow kernel of the coconut, stripped of its husk before being ground up and dried) were widely used as feed for cows, despite their significantly lower nutritional value (a nitrogen content of between 3 percent and 3.5 percent). As for the castor bean – the nitrogen content of which ranged from 5 percent to 6 percent in unhusked cakes, falling as low as 4 percent in cakes of untreated beans – its use as a fertiliser was fairly widespread on account of its low cost, despite the fact that handling it led to irritation of the mucus membranes due to its highly caustic nature; a non-edible seed, the castor bean originally produced an oil much appreciated for industrial purposes, with Marseilles representing a focal point of production; in farming it was used exclusively as a fertiliser, given that it would have proven toxic as a feed for livestock.

At first glance, some observations are possible primarily on the basis of the volumes exchanged. The fluctuations shown by the graphs can be explained through a combined study of various factors: first of all, the level of oilseed imports, or the factor underlying French oilseed cake production; secondly, consideration must be given to the sum total of French customs legislation, as well as the treaties that governed trade with the countries involved in the buying and selling of seedcakes. An initial point to be made is that France was, at one and the same time, a producer, a consumer and an importer of seedcakes. This situation arose due to two main factors: a) the noteworthy role of the oilseed industry, which drove the imports of seeds and used cakes as one of its important sources of income, along with other by-products (soap, food products, seed oils); b) the extensive variety of seedcakes available at different levels of nutritional value and price, giving rise to multiple import and export circuits. Available data often do not allow to disaggregate oilseed for feed or for manure.

The purpose of these graphs, as a first approach to the question, is only to suggest a general idea of the extent of the oilseed trade and the countries to which some productions were directed. The dynamics of foreign trade in their complexity are not the focus of this study. Their articulation and complexity are addressed in multiple studies that cannot be reported upon in detail.

Graph: Fig. 3 Foreign Trade of Cakes, Italy and France, 1919-1936 in Metric Tons. Source: For Italy see Fig. 1, for 1921 are listed "Panelle di noce o altre materie" and "Panelli di semi oleosi e di altre materie"; from 1921 onward "Panelli di semi oleosi e di alter materie". For France, I have elaborated data from Direction Général des Douanes, Tableau général du commerce et de la navigation, premier volume, Commerce de la France avec ses colonies et les puissances étrangères, Paris, for the years 1919-1939 (henceforth: Tableau). Note: For France, here and in figure 4 I have added all seedcakes ("tourteaux de graines oléagineuses", "autres tourteaux et drèches"). After 1934 "amurcas" and "grignons d'olive" are also listed. The data considered are that of "special" foreign trade (directed at the unloading of the goods and not re-exported). See S. Becuwe/B. Blancheton/K. Onfroy, Données du commerce extérieur de la France between 1836 and 1938 - Base Montesquieu, in: Cahiers du Gretha 35, 2015, p. 4.

Graph: Fig. 4 Import of Fertilisers in France, 1919-1939 in Metric Tons. Source: See Tableau (Fig. 3). Note: In the category of "organic fertilisers" are also listed "os calcinés à blanc"; "Noir d'os (Noir animal)" [in 1927 separated in three categories: without category, "lavé ou en pâte", "autre"; in 1928 only two categories: without specification, "lavé ou en pâte"; from 1929 "lavé ou en pâte", "autre"]; "oreillons"; "autres produits et dépouilles d'animaux".

In the case of France, the noteworthy growth in the volume of trade that immediately followed the First World War encountered an initial period of retrenchment in 1924-1925: this was on account of the inflationary and monetary developments that affected the Franc up until the stabilisation achieved under Poincaré in 1926, and then there were the variations in the export policy imposed by the Ministries of Agriculture, Finance and Trade (see below, Section 4). Still, it was not until the end of the 1920s and early 1930s that the sector underwent the most significant fluctuation, traceable to a number of different factors we could try to formulate:

• ‒ the saturation of the oilseed market that occurred in 1933, due to, among other things, a number of legislative measures; the threat of possible controls on the quantities of oilseed that could be imported, with the exception of those from colonial territories, alongside the law of August 6 th that placed an incoming customs duty on all seeds not arriving from the colonial territories, and drove importers to stockpile large quantities of seeds in their warehouses; thanks to these reserves, domestic production was kept at levels that met the needs of a farming sector in crisis, pushing down purchases from abroad from 1933 onward.
• ‒ the onset of the global crisis had a number of different repercussions: a) imports of chemical fertilisers, which were generally more expensive than oilseed cakes except phosphates, decreased also due to the drop in consumption (see Table 4); b) this probably set off a mechanism of partial replacement, in the form of increased imports of the (more economical) seedcakes in 1933, a trend running counter to the drop in imports of chemical fertilisers; immediately following this peak, however, came a steep drop-off that lasted until 1935 (Fig. 2); c) after the peak of 1930, and with the exception of the year 1935, exports of French seedcakes, which had a high value-added (as shown in section 1), declined.
• ‒ the political and customs-related developments of the 1930s contributed to a noteworthy change in the make-up and relative importance of the countries involved in the seedcake trade with France, as certain countries reduced their purchases in the 1930s (Germany, Sweden, Switzerland, Fig. 5).
• ‒ purchases from abroad dropped in 1934, in part due to the increased availability of domestic products. Lower prices for French secondary cereals made these products more competitive than foreign seedcakes, especially as livestock feed.

Tab. 4 Index of Use of Chemical Fertilisers, World, 1929–1934 (nutrient).

4 Source: Our elaborations on Concimi e concimazioni 6, 1936, p. 124, which quotes an article by A.N. Gray in the journal La technique agricole internationale. 1929=100. Note: Data on nitrogen also include chilean mineral nitrate and, differently from the other nutrients, are divided per crop-year (1929 is 1929-30 and so on).

Graph: Fig. 5 Import and Export of Seedcakes in percent (France, 1919-1936). Source: See Tableau (Fig. 3).

The study of the Marseilles case allows us to illuminate the complexity of the dynamics that contributed to determining the spread of oilseed cakes in the countryside, both as animal food and as a fertiliser. The period following the First World War began with the difficulty of selling off the stocks that had accumulated in warehouses. The chief concern of the Ministry of Agriculture was to ensure that seedcakes were available to use as feed for French livestock, hence their reluctance to reopen the market for sales abroad. The ban on exports that was still in effect – and was soon to be lifted in part, though never in lasting fashion – was the root cause of excess inventory held by the oilseed producers. The problem became all the more pressing as domestic demand proved slow to recover, especially during the summer months, when it was easier for livestock to find hay as their main source of nourishment. The ban on exports, decided in July 1919, was suspended in June 1920, shifting the focus of the conflict to a competition between individual local manufacturers seeking to procure quotas of the export market. An export duty was introduced, however, and not until June 1921 was a fully open system put in place. Then, in July of that same year, a new ban was issued by the Ministry of Agriculture. Scheduled to go into effect in the month of September, it drew the ire of the manufacturers and exporters of the Marseilles region.

An exchange of letters that took place in the spring of 1920 between the Syndicat de fabricants d'huile (Oil Manufacturers Union) of Marseilles, the city's Chamber of Commerce and the French Ministries of Trade and Agriculture in Paris provides a good picture of the situation.

On March 16th 1920, the Syndicat informed the Ministry of Agriculture that in the warehouses of the producers of the city there were 26,350 tons of edible oilseed cake and 6,800 tons of fertiliser ones, while the estimated annual production for 1920 would have reached 248,800 tons of edible oilseed cake and 66,100 tons of fertiliser cakes. To have an element of comparison, it is enough to remember that during the five years 1871-1875 the average of the stocks had been 10,320 tons, compared to over 30,000 in 1920 (Table 5); but 1871-1875 was a period marked by an increase in the production of cake, which however was not entirely absorbed by the local market and that was largely exported.

Tab. 5 Stock of Seedcakes: Oil Manufacturers Union of Marseilles, March 1920, in Metric Tons.

5 Source: Letter of the president of the Syndicat de Fabricants d'Huile of Marseilles, Émilien Rocca, to senator Donon, Marseilles, 8.04.1920, in: ACCIM, MP/3.7.2.2.2/01.

The city's leading oil manufacturers complained of significant product stocks remaining unsold, particularly during the summer. The president of the Union of oil producers, Émilien Rocca, made the case for setting up a different regime between winter, when production was substantially absorbed by French farmers, and summer, when the production of cake did not easily find outlets in the countryside, due to the greater presence of fodder. Hence there was the need to open the export trade in the summer, with a quota of about 7.000 tons per month which, according to Rocca, would not have penalized the supply of national agriculture. The rural world, on the other hand, had a different opinion, in particular the Ministry of Agriculture, which considered it a priority to maintain a stock of feeding oilseed cakes, without which it would have been necessary to import non-bread-making cereals.

Probably, the first to halt production at the beginning of 1922 were the Verminck seed oil plants, giving as the true cause of the impasse the impossibility of exporting their seedcakes abroad. The President of the Verminck establishments considered the obstacles to exports imposed by the government authorities to be responsible for the lost earnings of production and the difficulties in supplying raw materials. Furthermore, he indicated that the Italian oilseed cake industry was as a fierce competitor, particularly in the Swiss market, where Italy sold its cakes at a higher price than Marseilles producers could. Hence the decision to stop the production of the 100 tons per day of peanut oil cakes produced by the plant, starting from January 15th 1922.

The representatives of the city's manufacturers and exporters made their voice heard, and were supported by the Chamber of Commerce of Marseilles as well, which declared, in an official resolution of January 17th 1922: "It is truly unusual, as well as unfair, that, in order to provide French farmers with inexpensive seedcakes, and despite the fact that the farming sector fails to absorb the quantity of seedcakes we already produce, we of the oilseed industry are deprived of our foreign sales outlets, forcing us to offset this loss by raising the prices of our oils, at the expense of another category of consumers". It was also noted that attempts had been made to accommodate the interests of the farmers, by providing them with a certain quantity of seedcakes at prices set on the basis of the benchmark listings of the London exchange, only to have these overtures rejected by the Ministry of Agriculture. The result would have been higher sales prices for manufacturers of soap and, in general, for other consumers of oilseed. The clash between the farming sector and the world of industry could not have been sharper.

Ultimately, periods of free trade alternated with frequent moments in which exports were subject to fully-fledged shutdowns, as well as other periods in which exports of French seedcakes were governed by individual government licences issued in response to specific requests from manufacturers.

An agreement was reached in March 1922, guaranteeing that the needs of French farmers would be met with guaranteed stocks at prices set, of common accord, at lower rates (10 percent lower for Rufisque peanuts and 5 percent lower for the Coromandel variety). The operation was handled by the central Paris office of the Union des Fabricants d'Huile de France. The levels of the stocks and the discounted rate were modified on a number of occasions during the 1920s.

The Chamber of Commerce of Marseilles, in the words of Georges Oppermann, whom the Chamber had assigned to study a new legislative proposal calling for the export of seedcakes to be shut down completely, did not hesitate to refer to the years just passed as a period of "lutte terrible" against a Ministry of Agriculture that had "stubbornly" insisted on maintaining the ban on the export of seedcakes. In addition to this underlying conflict, which continued on and off throughout the first half of the 1920s, a further dispute broke out at the local level between the manufacturers and sellers of seedcakes leading to, among other outcomes, the establishment in 1924 of a regulated market, based on the example of what had been set up in Paris, Rouen and other major French cities for a variety of different products.

In short, the needs of the farming sector, as put forth by the Ministry of Agriculture, involved dealing with spiralling inflation and increases in the cost of living by providing farmers with supplies of livestock feed at costs lower than those found on the world market. The agreements reached in 1925 by the Ministries of Agriculture, Trade and the colonies served as a litmus test of this situation, in that they allowed unlimited export of Coromandel cakes (those not consumed by French farmers) while requiring preliminary authorisations for the other types of seedcakes, with the underlying condition that a minimum level of stocks be guaranteed to French farmers, and maintained at a maximum price no higher than 75 percent of the price set for foreign purchasers of the same product. A joint commission established under the auspices of the Ministry of Agriculture was to oversee the granting of authorisations, but the system was paralysed in October 1926, when disagreement between the manufacturers and the traders led to a block in the granting of the licences. The tormented sequence of openings and shutdowns to foreign trade continued during the 1930s. As noted, that decade witnessed both the peak and the low point of foreign trade in seedcakes.

For Italy, the figures for exports cannot be only attributed to sizeable volumes of production, but rather to scarce internal use, a situation bemoaned on numerous occasions locally and discussed among the central authorities as well. An example is the speech by Vittorino Vezzani at the Chamber of Deputies, on June 10th 1929: "And just think that in these years the hay-based supply purchased by the market is the least economic, since it gives the nutritional unit higher price than almost any other feed; moreover, our consumption of concentrated feeds (flours, oilseed cakes, etc.) is still so scarce that a large part of these normally goes abroad from our factories (oil mills) which could instead work much better for us".

In Italy, it was concern over posing competition to the production and export of olive oil that created a climate unfavourable to incentives for the commercialization and processing of oilseeds. Moreover, unlike France, Italy did not have a colonial territory from which to buy the oilseeds necessary for the production of oilseed cake. For the legislators, the priority was to maintain the production levels of olive oil. The noble Giuseppe Augusto Pavoncelli became the spokesman of this need in an intervention in the Chamber, pronounced on May 1st 1930, in which his emphasis on nationalistic stances on the one hand and on tending to conserve the sources of income of the large landed properties especially in the south on the other hand, left no room for doubt:

"[...] it will be opportune to declare once and for all that the interests of olive growing are prevalent to each other in the field of national oil production both in relation to the oilseed squeezing industry and to the trade of re-export of foreign olive oil which is purified in Italy. And it could not be otherwise for the supreme political, economic and social purposes of the Nation and for the complexity of the interests that are linked to olive growing".

Pavoncelli conceived a role for the oilseed cake industry, albeit ancillary and subordinate to that of olive processing:

"The oilseed industry, with regard to the production of edible oils, must be traced back to its function of integrating the oil requirements of the Nation and must not claim to want to lose the taste of olive oil to Italians, a typically healthy and typically Italian product. Moreover, not only does it have a vast field of action for the production of industrial oils and technical oils that are still imported from abroad, but can effectively collaborate with olive growing, transforming itself for the refining of lampas and creating multiple possibilities of agreement with the olive growers for a convenient exploitation of the pomace for the cake industry, very useful for an effective increase in the national livestock".

It must be said, however, that the measures aimed to protect the olive oil production and processing sectors were not new in the 1920s. The first to be affected was the production of oil from cotton seeds (1881) followed by a rule that discouraged the production of all seed oils. This was true to such an extent that the increased tariff rates on oilseeds and oils made from seeds, stipulated by Italy in 1929 (Ruling no. 2038 of December 3rd 1929), led to an increase in imports of ready-to-use seedcakes (Table 6). In this regard, it is important to remember that cakes from olive oil were of little value which, moreover, led to a higher price of olive oil compared to other vegetable oils. Also, for this reason, olive oil producers found the path to the production of oilseed cake difficult.

Tab. 6 Import of Oilseed Cakes: Italy, 1928-1936 in Tons.

6 Source : see Fig. 1 and 2.

The alleged separation, evoked by the congressman Pavoncelli in 1930, between primary oil production for food, and secondary production of industrial oils or oilseed cake was entirely theoretical. In industrial practice, the two markets were not so far apart and hindering a possible market outlet (production of edible oils from oil seeds), it ended up hindering the whole chain of seed processing, failing, however, to protect olive oil production, which had to cope with large quantities of imported edible oil. This dysfunction appeared very clearly after the end of World War II.

In France, the customs tariff of 1933, which affected the importation of oilseeds not coming from colonial territories, gave rise to an even more incisive regime of protection and intervention. Oilseed cakes also fell under this restriction: on December 28th 1933, the decree establishing the quota for importing oilseed cake and tax payment was approved. Despite protests by importers of non-food oilseed cakes, such as castor oil, destined for use as fertilisers, the Ministry of Agriculture maintained its positions. As the Minister himself wrote in a letter to the Chamber of Commerce of Marseilles, "the French oil industry is able to supply all the oilseed cakes that can be used in our agriculture". Once farmer demand was satisfied, priority was given to the protection of the oil industry, which should not have suffered competition from foreign oilseed cake.

In France, during the 1920s and 1930s, there was a substantial intervention by the ministerial authorities in defence of the rural world, tending to guarantee a low cost for feeding livestock. If the customs protection of 1933 gave guarantees to the oil and cake producers, the peasant world remained a constant thought of the authorities, who did not hesitate to introduce new export barriers in the event that national production reached critical moments. This is what happened on July 6th 1938, when new customs tariffs were imposed on exports, following a particularly dry period in 1936-37 which could have put the supply of fodder for livestock into difficulty.

The use of oilseed cake in Italian agriculture was less widespread and more recent than in the French or Northern European context. For the same reason, governmental sensitivity for this matter was also lower. Moreover, Italian breeders faced a context with greater distrust of oilseeds and related production, due to the greater economic and political weight of olive oil producers, who in the interwar period experienced severe crisis. The diffusion of oilseed cakes for livestock, a distinctive sign of the intensification of breeding that occurred all over Europe, albeit at different times and scales, took place at the crossroads of multiple dynamics establishing timing and methods. The spread of organic fertilisers seems to confirm that the modernisation of European agriculture played out across different fronts, all closely interrelated.

In the specific field of the production and commercialisation of fertilisers, it is possible to use the categories employed by Laurent Herment to describe the essence of agrarian capitalism: "it is downstream and upstream of the strictly agricultural production process that agrarian capitalism developed". It is possible to distinguish two different levels: on the one hand the factors that influenced the production, on the other hand the matters related to distribution. As for the first aspect (the terminus ante quem), it emerges that the agronomic logic, in itself, as important as it is, is not sufficient to explain the use or not, the success or otherwise of the oilseed cakes. As the Italian case shows, there was certainly no lack of experimentation nor of studies in the Peninsula, albeit somewhat later than in France. In order to explain the use of oilseed cakes to the agronomic component, it is necessary to add the industrial one, as well as the articulation of the interests linked to the existing agricultural systems (native rapeseed producer vs breeders in France, olive supply chain vs oilseed supply chain in Italy). On the other hand, the decisive role of the term post quem also emerges, namely the circuits linked to the commercialisation of the oilseed cakes produced. The commercial policy of the states, in this perspective, played an important role in the diffusion of some vectors of agricultural modernisation. The links between these aspects are very close, as shown by the pressures of some social groups on the policies of the states in the period analysed.

For comments on a preliminary version of this paper, I would like to thank Gianpiero Fumi, who directed me to relevant bibliographic corpus on oilseed cakes in Italy, Léo Charles, Laurent Herment, Arnaud Page, Christine Strottman and the participants of the panel 9 of the EURHO Conference 2017, held in Leuven. The responsability for the results is entirely mine.