After the Harvest

The beef was surrounded by jelly, with two pieces of fowl beside it. When the examiners warmed the contents, they found the soup good and the meat tender. The concentrated broth in another vessel seemed scarcely different from a fresh preparation, although Nicolas Appert said he had made it fifteen months earlier.

These judgements appeared in a report delivered to a French industrial society in March 1809. Its committee included the chemist Guyton de Morveau, the food specialist Antoine-Augustin Parmentier and Bouriat, who presented their findings. They examined milk, whey, peas, beans, fruit and juices as well as meat. Appert was offering something useful to a household, a hospital or a ship: familiar food available long after its ordinary season.

The committee also stated a limit. The samples had been in the society’s possession for only two months. For the longer interval between preparation and opening, its members had to accept Appert’s account. They could warm the soup, taste the vegetables and inspect the fruit. They could not independently account for every month of the food’s previous life.

The work before the boil

Appert had learnt preservation through food and drink. He described a working life spent among the pantries, breweries, cellars and storehouses of Champagne, then the premises of confectioners, distillers and grocers. He knew what the established methods did to their ingredients. Drying hardened them; salt altered their texture and flavour; enough sugar to keep fruit could overwhelm the taste he wanted to retain.

His alternative combined a closed container with heat. Food was enclosed, the opening carefully stopped, and the whole vessel heated in a water bath. It then cooled without being opened. The preparation and heating varied with the contents. The bottle could hold a portion of the year’s harvest without requiring it to become a pickle or a heavily sweetened preserve.

At Massy, near Paris, this became a manufacturing undertaking. The food historian Olivier Lepiller describes a workforce that grew to as many as fifty people, including nearly thirty women preparing vegetables. Their trimming and sorting belonged to the process just as much as the equipment did. The first-person instructions in Appert’s book cover work distributed through several rooms.

One room held the cooking equipment and a large broth kettle with a perforated inner vessel for meat and poultry. Another was assigned to milk, cream and whey. Corking, tying and bagging took place separately, before the containers reached a room with three large copper boilers. Other spaces accommodated fruit, vegetables and washed empty bottles.

Appert wanted these arrangements ready before processing began. Peas made the consequences of delay particularly plain. If picked too young, they could dissolve during treatment. If left for two or three days, they could lose flavour, harden or begin fermenting before reaching the bath. Some bottles broke during heating. Preserving the harvest began with deciding which part of it was fit to preserve.

What held the cork

An ordinary bottle did not necessarily suit the job. Appert complained that familiar bottles had narrow, badly made necks and glass too weak for both the corking blows and the heat. He had bottles made with wider openings and a projecting ring inside the neck. That internal narrowing gripped the compressed cork and helped it resist being driven upwards when the contents expanded.

The stopper had to be prepared too. A lever compressed most of its length, making it more supple and reducing the end that entered the neck. Appert insisted on fine cork. Saving a small amount on a stopper could lose the much more valuable food below it.

A wooden support held the bottle while the cork was driven in with a bat. Part of the cork remained above the opening, giving two crossed iron wires something to hold down. The wires were already secured when the bottle entered the bath. The process depended on retaining the closure while heat acted on the contents.

Large jars brought a different difficulty. Appert could not obtain the same internal ring in their broad mouths, and suitable cork was not always thick enough. He made composite stoppers from pieces joined with glue. The fitted stopper, an outer sealing compound and the wire restraint each had work to do; replacing a bottle with a larger jar was not simply a matter of increasing its volume.

The filled vessels went into linen sleeves before heating. Earlier packing with hay or straw had left broken glass to be picked out afterwards; a sleeve kept the fragments of a failed bottle together. Heating was followed by staged cooling, and then an inspection of each container. Cracks, star-shaped flaws or moisture around the stopper meant a bottle had to be set aside rather than stored with the sound ones.

Appert acknowledged that failures still occurred. A defective cork, careless corking, too much food in the vessel or badly secured wire could spoil the result. Knowing that the general method worked did not remove the need to judge each bottle.

The bottles go aboard

Naval trials carried the claim beyond Massy. A Brest Council of Health report from 1803 described provisions sent by the Minister of Marine and kept for three months in the roadstead, the sheltered anchorage off the port. Its five signatories found the bottled broth good. Broth accompanying the boiled meat was good but weak; the meat itself remained acceptable. Peas and beans retained the flavour of newly gathered vegetables.

Another consignment went aboard the Stationnaire in December 1806 and came ashore in April 1807. The published Brest report concerned eighteen preparations. It said they had undergone no change aboard and were in the same state when landed as when embarked. The evaluators also wanted a practical improvement: fewer containers. Feeding a ship with many separate bottles would create its own handling and storage problem.

An admiral’s letter offered a more domestic sort of testimony. Zacharie Allemand wrote from the Majestueux, anchored off the Île d’Aix, that his captains had tasted vegetables bought from Appert fourteen months earlier. His steward had accidentally left a bottle in the store-room. The officers took the peas and beans for fresh ones and wanted to buy more provisions. This was an endorsement from a customer, rather than a controlled trial, but it gave Appert an account in which preserved food had passed without announcing itself.

The Paris committee’s own examination was similarly attentive to the food. Cherries and quartered apricots retained much of their flavour; the fruit had been picked before full ripeness to help it keep its shape. Raspberry aroma and currant acidity survived, though the juices had faded in colour.

The committee had not yet been told Appert’s method. Its favourable report helped establish the result before the process became public. Government examination then went further: a Board of Arts and Manufactures observed experiments and later certified that Appert’s written description agreed with what he had demonstrated.

Two hundred copies were a condition of payment.

On 30 January 1810, Interior Minister Montalivet granted Appert 12,000 francs as a reward for the useful process and compensation for expenses. He required a detailed description, examined by the board and printed at Appert’s expense, with two hundred copies supplied to the ministry. The letter gave a financial reason for turning the manufacturer’s knowledge into a book others could use.

A different container, a different explanation

The process soon acquired other containers. Peter Durand’s British patent, dated August 1810, described food enclosed in glass, pottery, tin or other suitable materials and treated with heat. It explicitly acknowledged communication from someone abroad. Historical research identifies Philippe de Girard in that transfer; Durand’s name on the patent does not make the whole preservation method his independent invention.

In observations published with the specification in 1811, Durand described replacing glass vessels with tin cases and treating as much as thirty pounds of meat at once. He sent cases of meat, milk and soup to Joseph Banks to be opened before him, and reported inspections by members of scientific societies too. The larger package was accompanied by a familiar request: let respected people inspect food after time had passed.

Bryan Donkin and John Hall acquired the patent and developed commercial canning. Appert also went on working with metal. His enlarged book of 1831 included instructions for metal containers and work with an autoclave, a pressure vessel. The glass bottles of his earlier method were one stage in a changing technology.

His explanation of preservation changed less readily. He regarded heat as acting on the food’s tendency to decompose and considered exclusion of air essential. He had tested the combination well enough to produce persuasive results, without identifying the living agents responsible for spoilage. In the 1860s, Louis Pasteur’s work distinguished the air itself from the microorganisms carried in its dust; his bent-neck flasks could admit air while trapping that contamination.

Microbiology also exposes a limit to the early tasting reports. Heat-resistant bacterial spores can survive boiling-water treatment, and low-acid foods such as meat, milk and peas require a more severe process for dependable preservation. A sound seal and an agreeable flavour do not establish microbial safety. The commissioners’ reports tell us what the samples were like when examined; they cannot certify everything happening inside them.

For the customer, there was still the matter of dinner. Appert devoted instructions to using the food as well as storing it. He complained that if peas were poor in their season, the cook was blamed, whereas poor peas in winter were blamed on the preserver. Bad butter, rancid fat, careless cooking and a pan left too long on the fire could undo the earlier work. His own serving instructions ended with the peas piled on a thoroughly warmed dish.