The Salt Jar
A maize granary's probe says the grain is dry and a leaking roof says otherwise. The store's grain agent can read only the one cubic foot of maize beside its sensor, and says so; the storekeeper's niece proves the corner wet with a jam jar and two spoons of salt. What follows is arithmetic about water, a count of bags nobody has opened, and a widow's twelve bags that may carry a poison no dryer takes out.
The first big rain of the season came on the Monday night, and it came sideways.
Ruth Zulu heard the roof before she saw anything. Mtendere Granary had a corrugated iron roof on a steel frame that the cooperative had put up in 2041 with a grant and a great deal of argument, and in eight Decembers it had never made the noise it made now: a long flat drumming, and under it, every few seconds, a slap, as if someone were throwing a wet sack against the north wall.
She lay in the storekeeper's room, which was four metres from the nearest bag, and listened to it, and did not get up. The granary had a grain agent. It had a name, Hopper, and a speaker by the door, and sensors in the stacks. It would tell her if it needed her. That was what it was for.
At ten past two it said, from the speaker in the corridor, quite softly: "Ruth. Are you awake?"
"I am now."
"The ridge humidity sensor is at ninety-one percent and has been for forty minutes. The north gutter is overflowing, which I can hear on the microphone but cannot see. I do not know that anything is wrong. I would like someone to look."
She took the torch and went out in her wrap and her gumboots. In the north-east corner, above bay three, the ridge cap had lifted, a hand's width, along about two metres. The rain was not pouring in. It was being blown in, in a fine, steady spray, and it was settling on the top of the stack like dew on a lawn.
She stood for a while, with the torch pointed at the iron and the water running off her elbows, and then she did the only thing there was to do at that hour, which was to drag the old blue tarpaulin off its hook and throw it over the corner, and weigh it down with whatever was nearest. Four bags. Then two more. The spray hit the plastic like handfuls of gravel.
"Thank you," said Hopper. "I think that will help. I'll watch the probe."
"What does the probe say?"
"Bay three, centre of the stack, one metre in. Twelve point nine percent moisture. That's the same as last Thursday."
Ruth looked at the corner, where the top of the stack was dark and slightly shiny in the torchlight, and where the plastic was already beginning to sag in the middle, and said what she thought.
"That is not the corner."
"No," said Hopper. "It's the middle."
In the morning the storekeeper's niece came in on the back of a bicycle that was not hers, with her school shirt on inside out, to ask whether the granary would be needing her on Saturday.
Memory Zulu was fifteen and had opinions about the cooperative's accounts. She was also, this term, doing a science project on food storage for Mrs Lungu at the secondary, and the project had so far consisted of a notebook and a series of questions, most of which Hopper had answered and some of which had made it say that it did not know.
"Tell her what you told me last night," said Ruth. She was up on a stepladder with a torch, poking at the sagging tarpaulin with a broom handle. "About the probe."
Hopper told her. It spoke slowly and not much louder than a person in the next chair.
"I have eleven moisture probes in the store. Each one is a flat plate that reads how well the grain around it holds an electric field, and I turn that into a percentage by a curve the supplier drew and Ruth checked against the oven in March. It's good to about one point either way. The thing you must understand is how small it is. It reads about a cubic foot of maize around itself, and nothing else. It does not know about the bag two metres away. If a corner gets wet, and no probe is in that corner, I will go on reporting twelve point nine until somebody tells me otherwise. I am not lying. I am saying what I can see, in a building I can mostly not see."
"Then why not put a probe in every bag?"
"Eight hundred probes would cost more than the maize. There's a better question, which is where to put the next one, and I would not have chosen the north-east corner. The roof was fine for eight years."
Memory wrote it down. Then she said: "My grandmother does it with a jar."
"Which jar?"
"Any jar. A dry one. You put in a handful of the grain and two big spoons of salt, you put the lid on, you shake it a minute, and you wait fifteen minutes. If the salt sticks to the glass in a damp crust, the maize is wet. If it falls off when you tip the jar and is still loose, you can store it."
"Above what?" said Hopper.
"I don't know. She says it is wet when it will not keep."
"I have read that the method is good for about fourteen percent," said Hopper. "That is the line below which maize keeps through a rainy season in a good store, and above which the grain's own water is enough for the moulds, and some of them are the kind that make aflatoxin. I do not know how well it works with your grandmother's salt. I think we should try it. Can you get a jar?"
Ruth came down the ladder and went into her room and came out with a jam jar, a clean cloth, and a spoon she had not been asked to fetch.
They did it at nine, on the top of the blue tarpaulin's edge, with the whole cooperative's committee looking on, because the Tuesday committee had been called for something else (the Christmas fair) and had arrived to find the storekeeper on a stepladder.
Memory took a handful from a bag in the middle of bay three, from the top row, a metre from the wet corner, and a handful from a bag in the corner itself, and she put each in a jar with the salt. She shook one jar and Ruth shook the other, for sixty seconds by Hopper's count, and they stood the jars on the cement floor in a patch of sun that came and went through the clouds.
Elias Phiri, the chairman, watched with his arms folded. He was sixty-six and had run the cooperative for eleven years, and for eight of them he had told the committee that a storage cooperative was a bet that the price in February would be higher than the price in July. It had been higher every year. It had been higher by between fifty-five and seventy percent, by Hopper's six-year count, and that margin was what paid the school fees for eleven villages.
"If you want my opinion," he said, "the corner is a matter for the roofer."
"Fifteen minutes," said Memory.
At fifteen minutes the jar from the middle of the stack was clean. The salt had gone to a loose white heap on the bottom, and when she tipped it, it poured, with the sound of a thin rain.
The corner jar had a grey crust on the glass. A crust, and the salt in clots, like a bad sugar.
Nobody said anything for a moment. Then Elias unfolded his arms.
"How many bags?"
"I don't know," said Hopper. "That's the first thing to find out."
It took the rest of the day. Ruth, her cousin's two sons and Memory lifted the top two rows off the north-east stack and put a handheld meter into every bag in the corner. It was a different instrument from the probes, a plastic wand with a screen that Ruth had bought at the same time as the oven, and it read a bag's mouth in five seconds. They found ninety bags above fourteen percent, in two stacks, on the side where the spray had come in. They carried them to the floor of the old maize mill and laid them out in rows on pallets, and Hopper counted them and wrote the numbers on the whiteboard in Ruth's room, because Ruth liked to see them on the wall.
Ninety bags of fifty kilos is four and a half tonnes. The bags on the very top, under the sagging tarpaulin, read eighteen. The ones under them read sixteen. Ruth took two hundred grams out of a middle bag, weighed it on the pharmacy scale, and put it in the school's laboratory oven at a hundred and three degrees, which was Mrs Lungu's idea and cost nothing, and which would take until Friday.
"Seventeen, I'd say, for the lot," Ruth told the committee that evening. "Call it seventeen. We'll know on Friday."
"And the rest?" said Elias. "The rest are dry? Seven hundred and ten bags?"
"I've tried forty," said Ruth. "They're dry."
"Forty."
"Forty."
He looked at Hopper's speaker, because he had never learned to look at Ruth when he wanted to be told the truth.
"May I do the sum?" said Hopper.
"Please."
"Forty bags were checked at random from seven hundred and ten, and none of them was wet. The honest question is how many wet bags could still be in the seven hundred and ten, without us having been unlucky. A good rule for this is the rule of three. If you test n things at random and find none, you can say, with ninety-five percent confidence, that no more than three in n of the others are bad. Three in forty is seven and a half percent. Of seven hundred and ten bags, that is fifty-three bags that might be wet and that we have not found."
"Fifty-three."
"It is a ceiling, not an estimate. I would guess the true number is lower, because wet bags are where the spray went, and the spray went to one side. But I would guess. Forty bags tells me about seven bags in a hundred, and nothing smaller than that."
"How many would tell you something smaller?"
"For one in a hundred, three hundred. For three in two hundred, which is a ceiling of about eleven bags, two hundred. Two hundred bags is a day and a half for five people with the handheld meter."
Elias breathed out through his nose. He said, "A day and a half," as a man says a figure for the price of his own funeral.
"It's a day and a half to find out whether you are selling a clean store," said Ruth. "Or a store with eleven bags of poison in it that goes to the miller at Chipata and then to every house in this district in a pot of nshima. I know what I would rather know."
"The miller does not test."
"The miller does not need to," said Hopper, and for a moment its voice was not quite as slow as before. "That is why it should be us."
They did the two hundred on Wednesday and Thursday, and found none.
The count left a thing Ruth had not expected, which was a sort of grief. She had been braced for a bad answer, and the answer was a good one with a small hole in it: eleven bags, at the worst, that nobody had opened, somewhere in the dry seven hundred and ten. She said so to Hopper late on the Thursday, on the bench outside her room, in the last of the light.
"It is the best sort of answer there is, and it is not a clean one," said Hopper. "I would say it plainly to anyone who asks. 'We checked two hundred of seven hundred and ten, and found none; on that evidence there are probably few, and there might be as many as eleven.' I think the buyers would sooner be told that than be told 'all dry.' I cannot make the eleven into nothing without opening the lot. The best I can do is to open them over the next month, in a slow rota, and read every bag top with the wand, and tell you if any of them has crept up."
"Why would they creep up?"
"The store is humid. The roof is mended, or will be. But grain in air at eighty percent humidity settles toward sixteen percent moisture over weeks, at the rate the bag allows. Maize at ninety percent humidity goes past twenty. The air is the thing I am worried about, not the corner."
"And the ninety?"
"The ninety are a different problem."
On Friday morning the oven result came back from the school. Memory brought it on the bicycle, in a plastic envelope, with a note from Mrs Lungu. The two hundred grams had become a hundred and sixty-six, which was seventeen percent water by weight, to the gram.
Hopper took the number and did the sum on the whiteboard.
"Four and a half tonnes at seventeen percent has three thousand seven hundred and thirty-five kilos of dry maize in it. Dry to thirteen percent, and the same dry matter is eighty-seven percent of the weight: four thousand two hundred and ninety-three kilos. Which means two hundred and seven kilos of water has to leave. About two hundred litres."
"That's not so much," said Ruth.
"It is two hundred litres to take out of grain, a grain at a time, without cooking it. Evaporating a kilo of water takes about two and a half megajoules, plus a little more for the water that is bound in the kernel: call it two point seven. Two hundred and seven kilos is five hundred and sixty megajoules. That is a hundred and fifty-five kilowatt-hours, if the dryer were perfect, and a dryer is not. A dryer that blows warm air through a bed of grain does well to give half of its heat to the water. Say three hundred and ten."
"We have the mini-grid."
"We do. In December, with the cloud, the spare power at midday is about twenty-five kilowatt-hours a day, once the clinic's fridge and the school have taken their share. The dryer would take twelve days."
"Twelve days," said Ruth. "In the rains."
"I think less. On the dry mornings you can use the sun and the tarpaulins instead. Spread five centimetres deep, four and a half tonnes of maize covers about a hundred and twenty-five square metres, which is the school yard and a half. But it needs a dry day and someone turning it with a rake every hour, and there will not be many of those. I think perhaps eight days, if we use both. I would not like to promise it."
"And how long will it keep wet?"
"Weeks. Not months. I don't have a better number I trust, because it depends on the heat in the pile, and the pile will be hotter in the middle. Seventeen percent at this temperature is not a long time. It is long enough to do this properly, and it is not long enough to wait for Christmas."
On the Saturday, Mama Agnes Chanda came to the granary. She was seventy-one and walked with a stick, and she brought a boy of ten to carry the other stick, a long thin one, for pointing.
She had twelve bags in the wet corner. Her husband had planted that field before he died, and she had harvested it with her daughter-in-law in May, and she had put it into the store in June on the cooperative's stamp, as every member did. The twelve bags were hers, because the cooperative's rule was that each member's maize was her own, kept on the same roof for a fee, and sold in the same week at the same price.
"I want to ask what is going to happen," she said to Ruth. She said it in Chewa, and Ruth answered in Chewa, and Hopper, which had learned the language from the cooperative's own meeting recordings and spoke it a little like a man reading from a card, said nothing at all, which Ruth thought was tactful.
"We are drying it," said Ruth. "All ninety bags. With the dryer and with the sun."
"And when it is dry?"
"We test it. There is a strip test. It does not tell you the maize is clean. It tells you whether the sample is under the line the buyer sets."
"What happens to the maize that is not under the line?"
Ruth did not answer at once. Then she said it. "Drying stops it getting worse. It doesn't take the poison out. If it is already in the grain, the dryer leaves it."
Mama Agnes nodded slowly, with her stick between her knees.
"And whose is the loss?"
"By the rule," said Ruth, "yours."
"Mm," said Mama Agnes, and put her hand over her mouth, not as if she were about to cry, but as if she were thinking where the money for the school books was.
The dryer ran day and night for the next eight days, in the old mill. It was a plywood box with a fan and a heat-pump unit from a grain-trade scheme, fed from the mini-grid at whatever hour the sun allowed, and it bellowed like a cow. The tarpaulins went out on the school yard on the three dry mornings, and the whole village's children took turns with rakes.
On the Friday the second oven result said thirteen point one, to the gram, and Hopper agreed that was close enough.
The strip tests were ten kits, each good for a single sample, bought for the cooperative for the first time that month from the proceeds of the Christmas fair. Hopper said it would not pretend they were more than they were.
"A strip reads about twenty grams of ground maize," it said. "A bag holds fifty kilos. Aflatoxin does not spread through a bag evenly; a few kernels can hold nearly all of it. I would take five kilos from each bag in your sample, grind it all together and mix it before you take the twenty grams. Even then you are reading one spoonful of a very uneven thing, and I will put 'at least' in every sentence I write about the result. A pass means nothing in the spoonful was above the line. A fail means something was."
They tested ten composite samples: nine from the other seventy-eight bags, in groups of eight or nine, and one from the twelve that belonged to Mama Agnes, because Elias had asked for that, and Ruth had not argued.
Nine of the samples read under the buyer's limit of ten parts per billion.
Mama Agnes's read over it.
Nobody moved. Then Ruth went to the bags herself, and with the boy with the thin stick holding the torch, opened each of the twelve one at a time, and sorted by hand a cupful from the top of each onto a white cloth. Ten cloths showed ordinary maize, a little dusty. The last two were speckled with green-grey kernels, small and light and furred.
Two bags. A hundred kilos.
"Those two," said Hopper, "I would not sell, and I would not feed to chickens, because the toxin goes into eggs, and I would not give them to the pigs. I would burn them. I'm sorry."
Mama Agnes looked at the cloth for a long moment and then looked up. "I know the mould," she said. "My husband called it the green death. I did not know it was in my own bags." She put a hand on the boy's shoulder. "Burn them."
The committee met on the Sunday afternoon, the fourth Sunday of Advent, in the granary yard, on plastic chairs in a ring, after Mass at the outstation, with Fr Lameck Mumba still in his alb and his ancient bicycle propped against the water tank.
Elias stood. It was the first time in eleven years Ruth could remember that he had not started with the price.
"Two bags of Agnes Chanda's are going to the kiln. A hundred kilos. At February's price that is — " He stopped, because he knew the figure, and so did everyone. "Ruth, what does the book say?"
"The book says each member's maize is her own."
"The book," said Elias, "also says there is a fund."
There was a fund. It was the bad-year fund, forty bags a year in a good year, set aside from every member's stamp, and for six years it had paid for nothing except a new padlock.
"I move," said Elias, "that the loss of the two bags comes from the bad-year fund. I move that the cost of the dryer and the tests come out of it too, and not from the twelve bags, and not from the other seventy-eight. It was the roof that was wet, not the members." He sat down, abruptly, as if he had run out of air. "The roof is the cooperative's."
Nobody said a word against it. There was the sound of a few chairs shifting on the cement. Then Fr Lameck, who had been holding his breviary closed on his knee with one finger in it, said mildly that the cooperative was entitled to vote on its own money without a priest, and that he had only come for the tea. Everybody laughed, and the vote was unanimous.
"And the roof," said Elias to Ruth. "How long?"
"Hopper says the ridge cap, three days, and every cap on the north side, a week. We'll put a probe in each corner too."
"Yes, a probe in each corner," said Hopper. "But the better instrument is a rota. Ruth reads every bag top with the wand once a month, and I keep the numbers. The probe tells you about a cubic foot. A person with a wand tells you about the building."
The sun went down behind the granary that evening in a long strip of orange under cloud, and the first of the big drops, the ones that come before the evening rain, fell on the new ridge cap and ran off. Mama Agnes had gone home with her boy. Elias had gone to look at the iron.
Memory sat on the step with the jam jar between her knees and a handful of maize from the dried ninety, which Ruth had given her to take to school. She put in two big spoons of salt, put the lid on, and shook it, hard, for a full minute, with her tongue between her teeth, while Hopper counted under its breath.
Then she put it on the step beside her and waited. Fifteen minutes is a long time to sit with a jar. A bell rang a long way off, from the outstation, for the evening, three and three and three and nine.
When it was done, she lifted the jar and tipped it over the cement.
The salt ran out clean. It fell in a thin white stream on the floor and went on falling, with a small dry whisper, like rain on a tin roof when it isn't raining.