There was a paper sack by the screen door at the end of October, and the popcorn balls in it were still faintly warm through the wax paper. You handed them out to whoever came up the path. Their mothers had made something similar and sent it out of their own doors, and nobody involved thought this was remarkable, or dangerous, or worth writing down.
The video this page belongs to is a list of ten things that came out of ordinary American kitchens for pennies, and nine of them are gone. Popcorn balls at the door. Caramel apples you gave to other people’s children. Divinity at Christmas, which would not set if the day was humid. The taffy pull, which needed a second person and twenty minutes of an evening. Pudding stirred in a saucepan. Doughnuts out of a countertop fryer on a Saturday. Yogurt in a machine with seven little glass jars. Gorp mixed on the table to your own ratio. The wheat germ canister next to the flour. And Rice Krispie treats, which by every rule on that list should be as extinct as divinity and is being made in a kitchen tonight. Each one was killed by one of three things, and the video names them as it goes: the Box, the Clock, or the Card.
That is the video. This is the part that did not fit in it, and it begins with a hole in the middle of the argument.
Here is what I could not find. Nobody counted. There is no survey of how many American households made divinity in 1974, and none of how many make it now. No sales figures exist for a thing that was never sold. No participation data, no diary studies, no market research that reaches back far enough to be compared with anything. I went looking for a number that would let me say nine with a straight face, and it is not there, in any archive, for any of the ten.
So every use of the word gone on this page and in that video is a judgement rather than a measurement. It is built out of food writing, out of whether the products still exist to support making the thing at home, out of what turns up when you search and what does not. That is a reasonable way to reach an answer and it is not the same as knowing. The number in the title is my count of a list I chose, and a different list would have given a different number.
This page is everything I found while building that video, including the places where what I found argued with me. The sources are grouped under the claim each one belongs to, and the section at the bottom is the one to read if you read only one.
To see the video, go here.
The popcorn ball, and the wrapper that replaced it
The Smithsonian’s National Museum of American History dates the turn to the 1970s, and its wording is careful: parents became more suspicious of homemade items. (From homemade Halloween treats to colorfully packaged candies.) The museum says nothing about poison and nothing about sealed wrappers. That reasoning is mine, not theirs, and I should be clear about which is which.
The candy chemistry holds up. A popcorn ball needs syrup boiled to the hard-ball stage, which runs from about 250 to 266 degrees Fahrenheit, and that is a real threshold rather than folklore. (Candy stage temperatures.)
And the popcorn ball itself is not extinct. You can buy twenty-four of them, individually wrapped, this afternoon. (Popcorn balls, sold by the box.) That is the point rather than a hole in it. The object survived. The act of making two dozen and giving them to the neighbours did not.
Divinity, and the weather
Divinity fails above roughly 50 to 60 per cent humidity, because the sugar syrup pulls water out of the air and the candy will not set. Standard guidance is not to attempt it on a damp day. (Why humidity stops divinity setting.) That is why a grandmother could look out of the window and tell you whether Christmas candy was happening.
Food writers now describe it as a lost art, which is the closest thing to evidence of disappearance that exists for any item on the list. (The lost art of divinity.) Note what that is: coverage, not measurement. It is the best I have, and it is not a number.
The appliances that dated the decade
National Presto’s own corporate history gives the FryBaby in 1976 and the FryDaddy in 1977, with the GranPappy following in 1978. (National Presto company history.) Three fryers in three years is a company answering demand rather than inventing it.
Dunkin’ reached a thousand American shops in 1979 and opened its ten thousandth in October 2025. (1979, October 2025.) Both figures are United States only. If you see fourteen thousand quoted anywhere, that is the worldwide count and it does not belong in this comparison.
The yogurt maker in the museum is a Salton, catalogued around 1974, object 1982.3005.01. (Salton yogurt maker.)
Gorp, and how old it really is
REI’s own history of the word is the readable version. (The Origins of GORP.) The useful correction underneath it is that nuts and raisins carried in a pocket appear in camping manuals from the 1910s. (Trail mix, with the early history.) The bulk bins of the seventies were a stage in a long story rather than the beginning of one.
The wheat germ canister
Kretschmer belonged to International Multifoods in the seventies. The trademark record makes the registrant plain. (Kretschmer trademark, registrant International Multifoods.) Quaker Oats appears only in the eighties.
Quaker sold the brand to Brynwood Partners in November 2011. (Brynwood acquires Kretschmer, Baking Business on the sale.) The new owners then hired two agencies and built a website to explain wheat germ to Americans again. (The relaunch, King Fish Media and MyWheatGerm.com.)
Two things about that relaunch did not make the video. It was a superfood repositioning aimed at a new market, not a rescue mission for people who had forgotten. And America’s number one wheat germ brand is the buyers’ own description of what they had bought, not a measured share. Thirteen months later they sold the business on. (Continental Mills acquires it, December 2012.)
Wheat germ is also still stocked nationally. It never left the shelf. The claim is narrower and stranger than disappearance: nobody puts it in anything any more.
Rice Krispie treats, the one that lived
Invented in 1939 by Mildred Day and Malitta Jensen, working on a fundraiser. (Rice Krispies Treats, Who made the first one, the original recipe.) The recipe was on the packaging the following year.
The argument in the video survives the correction easily. A recipe printed on the side of a box that a company has every commercial reason to keep printing is a card that cannot be lost, and that is the only reason this one is still being made in kitchens while divinity is a magazine article about a lost art.
What did not survive
This is the section to read if you read one. Some of it is the video arguing with itself, and one item is a claim that would have been repeated for years if three separate checks had not killed it.
The patent that does not exist. I had a line saying trail mix was patented in 1968. It is a tidy fact and it is not a fact at all. The claim traces back to an unsourced blog sentence that has been copied ever since. Harmony Foods’ own account says 1969, their earliest trademark filing is 1978, and the food itself predates both by half a century. The line came out of the script entirely, and I would have believed it if nobody had gone looking.
The Halloween sentence that was falsifiable. The draft said researchers have never found a substantiated case of a stranger harming a child with Halloween candy. That is not what the research says and it can be disproved with one example. In 2000 a man in Minneapolis put needles in candy bars and a teenager was pricked. In 1964 Helen Pfeil handed out ant poison to children she thought too old to be trick-or-treating. Joel Best’s actual standard is narrower and it holds: no substantiated case of a child killed or seriously injured by a stranger’s Halloween treat. (Joel Best’s Halloween sadism research, Best and Horiuchi, The Razor Blade in the Apple, documented tampering cases.)
One more name belongs here, because it is the case people reach for. Ronald Clark O’Bryan poisoned a child with Halloween candy in 1974. He was the child’s father, and he did it for the insurance. He is the single most quoted proof of the stranger danger and he is evidence of the opposite.
The X-ray study proves less than it is used to prove. I said hospitals X-rayed 1,063 pieces of Halloween candy in 1988 and found nothing. Three things are wrong with that. 1988 is the publication year, and the screening ran in the mid-eighties around Reno. The 1,063 figure counts X-ray films, meaning bags, not individual pieces. And the finding was no radio-opaque foreign bodies, which means no metal. An X-ray cannot see poison, or powder, or most plastics. (Calvanese, Should we X-ray Halloween candy? Revisited, Calvanese 1986.) The study is still good evidence. It is evidence about razor blades and needles, and it has been carrying a much heavier load than that for forty years.
I will state a view here rather than leave it implied. The panic cost American children more than any poisoner did. It ended a thing that ran on neighbours trusting each other, and it did so on a fear that never found its case.
The museum’s missing four words. The Salton catalogue entry says the machine was among the most ubiquitous in households through the 1990’s. An early draft of mine dropped the last four words, which are the whole point of the sentence, because they say the appliance did not die with the decade that bought it.
The pudding was in a can. Hunt’s launched Snack Pack in 1968 in a single-serve metal can with a pull-ring lid. The plastic cup arrived in the mid-eighties. (Hunt’s Snack Pack, the canned pudding of the 1970s, period advertising.) Related: the refrigerated Swiss Miss pudding in the dairy case was a Beatrice Foods brand at the time, not a Hunt’s one. (Swiss Miss.)
1941 is the wrong year, twice over. The Rice Krispie treat recipe went on the package in 1940. 1941 is the year Kellogg’s renamed it to Rice Krispies Marshmallow Squares. And the pleasing claim that the recipe has been printed every year since, without a break, cannot be verified from anything I could find. It is very probably true. It is not established, and the video should not say it as though it were.
Yoplait was already here. General Mills bought the American rights in 1977, but the yogurt was being made in Michigan under licence from 1974. (General Mills on Yoplait, the Michigan licence.) Dannon’s In Soviet Georgia campaign is usually dated 1976 and Danone’s own account says 1977, so the safe form is the mid-seventies. (In Soviet Georgia.)
Twenty minutes was invented. A taffy pull takes about ten to fifteen. And the colour change I described, amber to white, only happens with sugar and saltwater taffy. Molasses taffy goes from dark and translucent to pale and opaque, which is a better sentence anyway. History Nebraska does keep taffy pulls in its archive, and their photograph is from 1914, so it is not a nineteenth-century image. (History Nebraska, Taffy Pull, the taffy pull as a vanished tradition.)
One source I could not open. The Calvanese X-ray abstract was read through two independent secondary sources quoting it near-identically, not through the paywalled original. If that number ever goes on screen as a graphic, somebody should buy the paper first.
And one thing I looked for and never found. No archived 1970s magazine advertisement for Kretschmer wheat germ. The script originally said it was advertised in national magazines. That may well be true and I cannot show it to you, so it came out.
A page like this goes up within minutes of each new video. Free, and it stays free.