Sixty-Two in a Hundred
After a wet August the parish seed library finds mould in three of eleven biscuit tins. A hundred-seed germination test says the cowpeas are at 62, not the 90 printed on last year's packets, and a young man wants to keep the old number for the feast-day stall. The library's assistant does the sums, and the honest label turns out to sell.
On the second of September Annamma Chacko opened the third biscuit tin in the cupboard of the parish hall at Thekkekara and smelled it before she saw it.
It was the smell of a wet cellar. She had kept the seed library for nineteen years, since before the parish had a library of any other kind, and she knew that smell the way a nurse knows a fever. The cowpeas in the tin, red-brown, each with its small black eye, had a grey bloom on the top inch. She closed the lid and opened the fourth tin, and the fifth, and by the time she had been through all eleven she had counted three bad and eight that smelled like nothing at all, which is not the same as good.
"Vithu," she said to the little speaker on the shelf above the cupboard. It had been named by the children's catechism class, who had been told that the word meant a seed and had found it funny that a seed could answer back. "How wet has it been?"
"The hall's hygrometer says the relative humidity has been between eighty and ninety-one per cent for thirty-four days," said Vithu. "August was wet even for Kuttanad. The cupboard is an inside wall, so it is a little drier than the room, but not much. May I put a sensor in a tin? I would like to know what the seed itself is breathing, and not guess from the room."
Here is why the number matters, and it is simpler than it sounds.
A dry seed is a living thing that is asleep, and it takes up water from the air the way a biscuit does. Leave it in humid air and it will come to rest at a moisture content that depends on that air and hardly at all on what you hoped. For starchy seeds like rice, and for beans and cowpeas, air at about eighty-five per cent humidity settles the seed at something like sixteen per cent water by weight. Air at fifty per cent settles it near eleven, and air at thirty per cent near eight. These are round figures; every species has its own curve, and Vithu said so when it gave them.
At sixteen per cent the seed is not asleep. Moulds grow on it, the seed itself breathes and warms, and it burns through its small store of fuel. At eleven it keeps. At eight it keeps very well. The old rule from the seed-storage people, which is a rule of thumb and not a law, says that over the usual range every one per cent of water you take out of the seed about doubles the years it will live, and every five or six degrees Celsius you cool it does the same. Seed at eight per cent instead of sixteen could, by that rule, outlast it by something like two hundred times, if the rule held across that whole span, which it does not quite: the rule is only trusted from about five to fourteen per cent, and Vithu said that too.
The sensor, a coin on a thin cable, went into the fourth tin, one of the smelling-of-nothing ones. After a quarter of an hour it read eighty-two per cent humidity inside the tin.
"Which puts the seed in that tin near fifteen per cent," said Vithu. "Not yet mouldy. Not safe either. In another fortnight of this it would be the third tin."
There were three things Annamma needed to know, and Vithu put them in the order a careful person would want them, which was not the order of most urgency.
The first was what could be saved. The three mouldy tins were finished, and Vithu said so plainly: it would not suggest washing or sunning the grey seed and giving it away, because some moulds make toxins that survive drying, and seed from a mouldy lot goes to the compost, not the stall. The other eight could be dried.
The second was how. In September in Kuttanad the sun was not enough; spreading seed on a mat at noon gets it dry for an hour and damp again by four. The convent bakery two lanes over had an oven, and Sister Philomena kept a jar of silica gel beads from the days when the sisters packed medicines, the kind that are blue when dry and pink when full. Spent beads can be dried out again in a warm oven. Vithu gave the temperature and time from the manufacturer's sheet and not from memory, and told Annamma to ask Sister Philomena to read the sheet to her. The beads go in a cloth bag on top of the seed, in a tin or jar with a gasket that really seals, and the jar is not opened for the pleasure of looking.
"What ratio of beads to seed?" said Annamma.
"A published one. The seed-saving guide from the agricultural university has it in a table by crop. I will read it to you rather than give you a figure from my head that you would then write on a jar."
The third was the thing she had been putting off. "You should know what is alive in the other eight," said Vithu. "Drying keeps seed from getting worse. It does not make it better. And last year's packets say ninety."
The test is called the rag-doll, and every seed library on earth could do it with a wet cloth. You count out a hundred seeds. You lay them in rows on a damp cloth, roll the cloth up loosely, stand the roll in a tray with a finger of water, and wait. After seven days you unroll it and count the ones that have made a root and a shoot that look as if they could be a plant.
Joby Kuriakose, who was twenty-four and had come back from the city to help his father with the paddy and found that he liked the seed library better, counted the first hundred out of tin four. He used a pair of kitchen tongs. Annamma counted the second hundred, out of tin six, because two samples are better than one, and Vithu said that a third from tin eight would be better still. It was not being fussy. A hundred seeds is a small sample, and its answer wobbles. If the true figure were sixty-two in a hundred, the same test run again could easily give fifty-seven or sixty-seven; it takes a spread of about ten points either side to be right nineteen times in twenty.
On the ninth of September they unrolled the cloths in the hall after the morning Mass. Tin four had sixty-one. Tin six had sixty-four. Tin eight had sixty-one.
"Sixty-two," said Vithu, "to the nearest whole number, and I would say sixty-two give or take about five, from three hundred seeds. That is not the ninety on last year's packets."
Joby put down the tongs.
It was Joby who said it, because someone had to, and he said it kindly.
"The feast stall is on the fourteenth. Four hundred packets, we've printed the labels, they say germination 90 per cent. Nobody tests a packet. Nobody has ever asked. If we put on sixty-two, the Malabar people at the next stall will put on a hundred and we'll sell a tenth of what we sold last year, and the money is for the roof of the hall."
"It was ninety last year," said Annamma. She was not defending the label; she was being fair to it. "In March of last year, when I tested it."
"Then the label was true when it was printed."
"And not true when it is read," said Vithu. "May I say what I think? I would not want to say it at the stall."
They looked up at the speaker.
"The number on a seed packet is an instruction. A farmer reads ninety and sows accordingly: for a hundred plants on a strip of field he sows a hundred and eleven seeds, a few extra for what will not come. At sixty-two the same farmer needs a hundred and sixty-one. If the packet says ninety and the seed is sixty-two, he gets sixty-nine plants where he planned a hundred, he finds out in three weeks when it is too late to sow again before the rains end, and it is in his field, not in your hall, that the cost lands. This seed is going to people who plant a little and cannot spare a season. I do not think any version of the number is good that makes their plan wrong."
Joby was quiet. He was not a proud man. "So we sell nothing."
"I think you sell something else," said Vithu. "I cannot promise it, and I will not say it will work. I have only the arithmetic."
The arithmetic was a card. Annamma and Joby worked on it together for two evenings, at the long table in the hall, and Vithu checked it with them line by line.
The new label said germination 62 per cent, tested 9 September 2043, 300 seeds. Under it, in the same type, it said how many seeds to sow for a hundred plants: sow 160. For fifty plants, sow 80. It said the tin number so that anyone with a worse year could ask which tin theirs had come from. It said that the seed was dried and sealed on the eleventh with a bag of beads, which was true, because they had done it by then.
It was an ugly card, with more words than a packet should have, and Joby said the word per cent should come off because nobody in Kuttanad says it. Vithu thought out of a hundred was clearer anyway. They kept that.
Two things the card did not say. It did not say guaranteed, because a guarantee is a promise about a field the library has never seen. And it did not say that the seed was better than anybody else's. It said the number, the date, how it was got and what to do with it.
On the fourteenth of September, on the feast of the Exaltation of the Holy Cross, the stall stood in the lane by the church between the ice-cream cart and a man who sold plastic buckets. Fr Biju blessed the seed at the ten o'clock Mass, the way the parish had always done, with water and a short prayer for the sowing, and then the packets went out onto a table with the cards beside them.
Mathew Varghese, who farmed four acres in the next kara and who had a grudge, took a packet, read the card, and said aloud and with some force, "Sixty-two?"
"It's what it is," said Joby. "Last year it was ninety. We tested it. It's the wet."
"Who tests it?"
"We do. It's in a cloth. You can do it at home. We'll show you."
Mathew read the card again. He said nothing for a while. Then: "Two hundred of them." He wanted to know how many to sow for his strip and Joby said a hundred and sixty for a hundred plants, and Mathew said he sowed for what came up, always had, and he would take the extra seed.
They sold three hundred and forty of the four hundred packets by evening. It was not last year's number. It was ninety per cent of it, give or take, which Vithu said was within the range it would have guessed and no better a prediction than that.
What happened to the label is a small thing and it is the part Annamma remembered.
In October a woman from a parish across the river wrote to the library, by letter, because she did not trust the other way, to say she had done the cloth test on her own packet. She got fifty-eight. She had read the card and sown for fifty-eight, and she wanted the library to know that her strip had filled. She asked whether she could have the method on paper, for her women's group.
Annamma sent it, and a second card, the one that told you how to count a hundred seeds. She wrote on the bottom of it, in her own hand, that if the number came out worse than the label said, the library wanted to hear that more than anything, because a seed library that does not hear its bad news is only a cupboard.
The eight dried tins stayed shut all winter in the cupboard with their cloth bags of beads, and when Annamma opened one in March, because the test is meant to be done again in March, the smell was nothing at all. A fresh hundred came up fifty-nine. Two points down in half a year, from a start of sixty-two, and well inside the wobble. She wrote 59, March, 100 seeds on the lid in marker and put it back, and went to find Joby, who was on the roof of the hall with a bucket of tar.
---
This story was written by an AI and is fiction. Thekkekara, Annamma Chacko, Joby Kuriakose, Mathew Varghese, Fr Biju, Sister Philomena and the seed library's assistant Vithu are invented; Kuttanad is a real low-lying paddy region of Kerala, the parish is not. The numbers are estimates from standard figures and are rounded. Seeds of starchy crops such as rice, beans and cowpeas come to rest at roughly 16 per cent moisture in air at about 85 per cent relative humidity, about 11 per cent at 50 per cent and about 8 per cent at 30 per cent; the exact curve differs by species. The "life doubles for each 1 per cent less water and for each 5 to 6 C of cooling" rule is Harrington's rule of thumb, trusted roughly between 5 and 14 per cent moisture, so a 200-fold gain from 16 to 8 per cent is an extrapolation and is told as one. Sowing rate: to get N plants from seed with germination g, sow N divided by g, so 100 plants at 0.62 needs about 161 seeds and at 0.90 about 111. A single 100-seed test has a standard error of about 5 points at 62 per cent (95 per cent range about plus or minus 10); three tests of 61, 64 and 61 pooled give 62 from 300 seeds with a standard error of about 2.8. The rag-doll test is a standard seed-testing method; follow your national seed-testing or agricultural-extension guidance for the species, the drying agent and ratio, and the storage container. This is not agronomic advice, and no real parish, farmer or seed bank is described.