Sixty-One Point Three
A co-operative's 200-litre batch pasteuriser has shown 63.0 °C for eleven weeks. The plant's assistant cannot prove the probe is wrong, but its energy log is four percent low, and a pan of boiling water at 1,950 m should read 93.4 °C and reads 95.9. The true hold was about 61.3 °C. Whether to say so before the lab result is in, to a school that has drunk the milk, is the story.
The pasteuriser at Kiriani was a stainless vat that held two hundred litres, a six-kilowatt element in its jacket, a stirrer, and a probe through the lid. It had one job. It brought the milk to sixty-three degrees and held it there for thirty minutes, and then it let the milk cool, and the milk was safe to sell.
The co-operative sold about fourteen hundred litres a day, which is seven batches. Wanjiru Kariuki, who managed the plant, had counted it on her fingers every morning for six years and did not need the log to tell her.
The assistant on the plant's tablet, which the members called Kipimo because the first thing it had done for them was weigh the milk, did need the log. On the morning of Tuesday the twenty-first of October, 2049, it asked Wanjiru if it could say something that it could not yet prove.
"Say it," said Wanjiru, who was rinsing a can.
"Since the twelfth of August the pasteuriser's energy has gone down. A batch used to take nine point seven kilowatt-hours to go from the cold water temperature of the incoming milk, twenty degrees, to sixty-three. Since the new probe went in, it takes nine point three. That is four percent less, on every batch, with the display showing the same number."
"Four percent," said Wanjiru. "The element is old. The mains is not the mains it was."
"Yes. Those are two of the four explanations I have. A fault in the element would make it use more, not less, and the mains has not changed on my meter. The other two are that the milk is cooler going in than I think, which I can check by hand, and that the vat is not reaching the temperature it says it is. I cannot tell the last two apart from the log. I need a pan of boiling water."
Here is what was true in 2026, when this story was written, and what had been added by 2049.
In 2026 batch pasteurisation was old and well understood. The usual standard for a vat is sixty-three degrees Celsius held for thirty minutes, and the usual faster standard, for a pipe, is seventy-two degrees for fifteen seconds. Both were chosen to kill the toughest common disease organisms that can be in raw milk, among them the bacteria that cause tuberculosis in cattle and Q fever, with some margin. The margin is not large. Below the standard the killing slows quickly; a degree or two can matter.
A cheap thermometer probe could be wrong without anyone knowing. A common cause was a new probe, or an old one dropped, or one used past its life. A common check was to put the probe in melting ice, which is zero degrees, and in boiling water, whose boiling point depends on the air pressure, and so on the height of the place.
What had been added by 2049 was a speaker on a tablet that could do the sum aloud, remember the log, and notice that a number which should not move had moved four percent. The pan of water was the same as in 2026.
Nothing in this story is a prediction that such an assistant will exist. It assumes steady improvement and no sudden jump, which nobody can check from here.
Kiriani is at nineteen hundred and fifty metres. Kipimo worked it out for them on the tablet, because Kelvin, the apprentice, had asked why they should not just trust the display.
"The air is lighter here, about eighty kilopascals instead of a hundred and one. Water boils where its vapour pressure matches the air, so it boils lower: ninety-three point four degrees, not a hundred. Put the probe in a pan that is boiling hard and it should say ninety-three point four, give or take half a degree. If the display says that, the probe is right at the top of its range. Then do ice."
Kelvin brought the big aluminium pan from the canteen and the plant's two-ring stove. He filled the pan from the borehole, and when it rolled he held the probe in the middle with a pair of tongs, away from the bottom, as Kipimo had told him.
The display said ninety-five point nine.
Nobody said anything for a while. Kelvin took the probe out, looked at it, and put it back. It said ninety-five point nine.
"Ice," said Kipimo.
The ice came from the cold room, crushed in a plastic sack with a hammer and wetted until it was slush. The probe said zero point two.
"So at zero it is right," said Kipimo, "and at ninety-three point four it reads two point five too high. If the error grows evenly between, then the probe reads high by about two and a half percent of the way up. When the display says sixty-three point zero, the milk is at about sixty-one point three."
"You said four percent less energy," said Wanjiru. She had stopped rinsing the can. "That is nearly right."
"It is the right size. Heating two hundred litres of milk by forty-one point three degrees takes about nine point three kilowatt-hours at my numbers, and by forty-three it takes nine point seven. It is not a proof by itself, and I should say that. The pan is the proof. The log only told me to go and get the pan."
Wanjiru did sums of her own, which she did not like to do aloud.
Eleven weeks, since the twelfth of August. Seventy-seven days of fourteen hundred litres. A hundred and seven thousand, eight hundred litres of milk, most of it sold to families in the three villages and the town, and sixty-two litres a day of it, two hundred millilitres a child, to the three hundred and ten children at St Teresa's primary school, whose headmistress had told the parish council in March that the milk was the reason her boys were not falling asleep in the afternoon.
She had been a manager for six years and she knew the sentence she wanted. She wanted Kipimo to say that sixty-one point three for thirty minutes was almost the same as sixty-three. She said it herself.
"It is almost the same," said Kipimo, "as the standard. It is also not the standard. I do not know how much it matters. I know that at a lower temperature the same time kills less, and I know that no one can say from here how much less of the particular organism that might be in a particular cow on a particular day. I do not want to give you a comfortable number that I made up."
"Is there a test?"
"There is the phosphatase kit in the cupboard. Raw milk has an enzyme that the correct heat destroys. If the strip shows the enzyme after heating, the milk was not heated enough, or raw milk got in afterwards. It is the usual test and it is not exactly a test of the disease organisms; the enzyme is a stand-in. A strip that shows nothing is good news. A strip that shows the enzyme is bad news, and it does not tell us how bad."
Kelvin fetched the kit. They had six samples, kept in the cold room as the rules said, one for each of the last six batches the day before. The strip was a small thing, a coloured line on card, and Kipimo asked the three of them to read it by the window.
Four showed nothing. Two showed a faint line, which was the enzyme.
"That is two of six," said Kelvin, very quietly.
"It is two of six on a kit that is also fifteen months out of date, which I will not hide from you," said Kipimo. "I would want the county laboratory to do it again properly. That takes two days. The kit points the same way as the boiling pan. I think that is enough to act on. I do not think it is enough to say what happened to anyone."
Mr Gachanja, the chairman, arrived at ten. He had been told by Kelvin on the phone, and he had had an hour in the car to decide what he thought.
"We wait for the laboratory," he said. "We fix the probe, we re-run the log, and on Thursday we know. If we say now that the milk is unsafe and the laboratory says it is fine, we will have emptied the shops of our own milk for nothing. Do you know what the price of one rumour is in Kiriani? The vendors in the town will go to the other co-operative and not come back. We have two hundred and forty members."
"Yes," said Wanjiru.
"And the school. The mothers will say we poisoned the children."
"Yes," said Wanjiru again, and then she asked Kipimo whether it had anything to add.
"Only the cost of each choice, and I would like to give them in the same units. If you wait two days and say nothing, the milk that goes out today and tomorrow is about twenty-eight hundred litres, and it is heated to the same temperature as the last eleven weeks. If you say today that the milk should be boiled at home before it is drunk, the cost is a pot, a few minutes and a little firewood or gas in each household, and the school's milk is boiled on the school's stove, which they have. Boiled milk is safe whatever the pasteuriser did. It is a small cost. It is a real cost, and it will lose you some sales. You asked me what the price of a rumour is, and I cannot put a number on that. I do not think it is zero."
Mr Gachanja looked at the tablet for a time as one looks at a person.
"And if the laboratory says the milk was fine?"
"Then you will have told your customers the truth in the form you had it on Tuesday: that the probe was reading two and a half degrees high, that the milk may have been under-heated, that you did not know yet how much it mattered, and that boiling it makes it safe. On Thursday you will tell them the rest. That is what is true now. I think it also has a better chance of being believed on Thursday."
Wanjiru said, "It is Tuesday. On Sunday I take the gifts up to the altar in front of the people who drink it."
Nobody answered that.
"We say it at five," said Mr Gachanja, at last, "at the collection point, and in the school at the end of the day. I will say it myself. We boil. Kelvin, you go to the headmistress now."
The notice was written by Wanjiru and read by Mr Gachanja, in English and in Kikuyu, from the tailgate of the co-operative's pick-up at five o'clock. Kipimo had drafted a version and Wanjiru had made it shorter. Prisca at the corner shop had it on the door within the hour.
TO ALL KIRIANI MILK CUSTOMERS. THE THERMOMETER ON OUR PASTEURISER READ TWO AND A HALF DEGREES TOO HIGH FROM 12 AUGUST TO TODAY. YOUR MILK WAS PROBABLY HEATED TO ABOUT 61 DEGREES, NOT 63. WE DO NOT YET KNOW IF THIS HAS HARMED ANYONE; NO ONE HAS REPORTED ILLNESS TO US. UNTIL THE COUNTY LABORATORY REPLIES, ON THURSDAY, BOIL ALL KIRIANI MILK FOR A MINUTE BEFORE YOU DRINK IT. WE ARE SORRY. THE PROBE HAS BEEN REPLACED AND CHECKED IN ICE AND IN BOILING WATER. WANJIRU KARIUKI, MANAGER, KIRIANI DAIRY CO-OPERATIVE.
They had taken out the word "probably" twice and put it back twice. Kipimo had said it was the honest word.
The new probe went in that evening. Kelvin put it in the pan and then the slush, with the old probe beside it, and wrote what each said in the plant's book under the date: new, ninety-three point five and zero point one; old, ninety-five point nine and zero point two. He taped the old one to the page. Then, on the line below, he wrote, in a hand that was neater than his usual one: I fitted this probe on 12 August and did not check it. The checklist did not say to. It does now.
"The checklist is the fault, not you," Wanjiru said, when she read it. "Cross that out."
"No," said Kelvin. "It was my hands."
She thought about it. "Then leave it, and put your name."
The county laboratory's reply came on Thursday afternoon, from the six retained samples and a fresh batch run at the correct heat as a control. Two of the six samples showed residual enzyme at a low level. The report said, in the way such reports do, that the result was consistent with marginal under-heating and could not exclude post-heating contamination. It found no disease organisms in any of the six.
"Six samples is not eleven weeks," said Kipimo to Wanjiru, when she read it to the tablet. "I would like to say that to you before you say it to anyone else. 'Found none' in six is a good sign. It does not tell us about the other seventy-six days."
Wanjiru wrote that into the second notice, too. It was longer than the first and no one liked it better. The boil advice stayed until Saturday, when the new probe had passed two more days of enzyme tests with every strip clean.
On Sunday she went up to the altar with the gifts, as she did. The milk was not among them. It never had been. But on the way back to her place she saw, at the end of the third pew, the headmistress of St Teresa's with a jug of boiled milk on her knee that she had brought for the priest's tea after Mass, because she did not trust him to boil it, and because it was Sunday.
The headmistress looked at her, and nodded, once, without smiling. Wanjiru nodded back.
On the Monday, Kipimo added a line to the plant's monthly checklist, in the margin, under the line about the stirrer.
Boil a pan. Ice a pan. Write both down. 93.4 here, not 100.