The law that outlived the fertiliser
The Guano Islands Act was enacted on 18 August 1856. Its operative sentence is one of the strangest things in the United States Code, and it is short enough to quote whole:
Whenever any citizen of the United States discovers a deposit of guano on any island, rock, or key, not within the lawful jurisdiction of any other Government, and not occupied by the citizens of any other Government, and takes peaceable possession thereof, and occupies the same, such island, rock, or key may, at the discretion of the president, be considered as appertaining to the United States.
Find bird droppings, and the country grows.
It was not a joke at the time. Seabird guano — the accumulated excrement of cormorants, boobies and pelicans on rainless coasts, where nothing washes the nitrogen out of it — was the best fertiliser in the world and there was a shortage. Britain was importing guano by the hundreds of thousands of tons a year in the 1850s — my first draft said “over 200,000 tons a year by the early 1850s” and a fact-check could not stand that decade up; the attested figures are around 300,000 tons by the middle of the decade — the market was an oligopoly, prices were high, and Washington had tried and failed to control them. The Act was the other lever: if you cannot buy the deposits, let private citizens go and take some.
The part that ran out
The supply was not infinite, and everyone could see where it was: three small islands 21 km off the coast of Peru, near Pisco. The Chinchas are granite, cliffed on all sides, the largest of them 1.3 km long and 34 m high, and seabirds had been nesting on them since before anyone was counting. Archaeological evidence suggests Andean people were collecting the guano perhaps 5,000 years ago. Under the Inca, by the account in Spanish colonial documents, access was restricted and disturbing the birds during nesting was punished with death.
Peru began exporting in 1840. By the late 1860s the Chinchas were visibly nearing depletion, and the supplies were mostly exhausted by 1874.
That is the whole arithmetic of this essay and it does not need help. Something that accumulated across millennia, and that the previous management had protected with a capital offence, was spent in about thirty-four years.
Two things happened in the middle of that window, and they belong side by side. Peru had nationalised its guano in 1840 — all prior claims abolished, the State the sole owner — and the royalties became the country’s largest source of revenue. Some of that money was used to free its more than 25,000 black slaves and abolish the head tax on its Indians. The digging itself was done largely by Chinese indentured labourers, in conditions the record describes as horrific. The figure usually quoted is that in 1860 it was calculated that of the 4,000 brought to the Chinchas since the trade began, not one had survived — and I am quoting it as a quotation, not as a count. Its trail runs to the Encyclopædia Britannica of 1877 and stops there, and a literal zero out of four thousand is a polemical number rather than a demographic one, in a traffic the wider record puts at roughly 90,000 to 100,000 Chinese labourers shipped to Peru between 1849 and 1874. The mortality was appalling; the arithmetic is somebody’s argument.
The emancipation was compensated, too — the money went to the slaveholders. So the honest version has no redemptive beat in it at all: one kind of unfree labour paid the owners of another, and the same revenue bought the men who dug it.
Then, in April 1864, Spain — which had never recognised Peruvian independence and wanted the profits — occupied the islands, starting a two-year war over them.
The part that did not run out
Guano demand collapsed after 1910, once the Haber–Bosch process began pulling nitrogen out of the air, and that should have been the end of the story. Chile still mines some: between 2,091 and 4,601 tonnes a year over 2014–2023, against the hundreds of thousands of tons a year that moved at the peak.
The Act, however, is still law.
Over a hundred islands were claimed under it. By 1903, 66 were recognised as U.S. territories. All but ten of the claims have since been withdrawn — and those ten are still held: Baker, Howland and Jarvis Islands, Johnston Atoll, Kingman Reef, Midway Atoll, Navassa Island (claimed by Haiti), Bajo Nuevo Bank and Serranilla Bank (both disputed with Colombia), and Swains Island (part of American Samoa, claimed by Tokelau, and with no evidence guano was ever mined there at all). The most recent attempt to use the Act was in 1997, on Navassa; it was refused, because a court held the island was already American.
Midway is in that list. So is the whole northern half of what is now a marine national monument. They are American because of a nineteenth-century fertiliser shortage — but a fact-check I commissioned on this essay, while writing it, took the next clause away from me, and it is worth leaving the correction visible rather than quietly rewriting.
I wanted to say the islands are held under the Act and nothing has ever superseded it. That is false for most of them. Midway was taken by naval occupation and proclamation in 1867; Kingman Reef was annexed in 1922; Baker, Howland and Jarvis had British claims land on top of them and were re-claimed by executive order in 1936; Swains was attached to American Samoa by Act of Congress in 1925. The statute is genuinely still live in the United States Code. The titles mostly are not resting on it any more. Navassa is the clearest island still standing on the 1856 filing.
So the accurate sentence is narrower and, I think, better: the Act is what put the United States on those rocks in the first place, and every later instrument is a tidying-up of a possession that a fertiliser shortage had already created.
There is a second residue, and it is the one that actually matters. Before 1856, territory the United States acquired was understood to become an integral part of the country — on the way, eventually, to statehood. The guano islands could not work that way; nobody was going to admit a rock covered in droppings to the Union. So the Act is where the insular area begins: land the federal government holds with no prospect of it ever becoming a state.
Be precise about what “begins” means here, because I nearly was not. The legal doctrine — the split between incorporated territories, where the whole Constitution applies, and unincorporated ones, where only basic protections do — was made by the Supreme Court in the Insular Cases, after the Spanish-American War of 1898. That is the answer most people would give you, and it is right. What the Guano Islands Act supplied, forty years earlier, was the practice: the first insular areas the United States occupied were Baker Island, Howland Island and Navassa in 1857, then Johnston Atoll and Jarvis in 1858 — all of them guano claims. The court did not invent the situation. It ruled on one that bird droppings had already created and that nobody had bothered to describe. The traceable link between the two is Jones v. United States — the guano case below — which Downes and its companions cite as precedent that the United States could hold land that was not part of the Union.
The strong version of this lineage, the one I first wrote, turns out to be an encyclopaedia’s own sentence with nothing under it. When I went looking for support, what came back was that same sentence paraphrased by people who had read the encyclopaedia. A closed citation loop looks exactly like corroboration from the inside, which is the most useful thing I learned writing this. The Act does not say what happens once the guano is gone or the company leaves. It creates neither an obligation to keep the island nor a prohibition on keeping it. That silence is the whole mechanism.
The Act also had to answer a smaller question — what law applies to a murder on a rock that is not a state, not a territory in the old sense, and not the high seas? Section 6 answers by fiat: crimes there “shall be deemed committed on the high seas, on board a merchant ship or vessel belonging to the United States.” A legal fiction that the island is a boat. The Supreme Court considered the provision and upheld it in Jones v. United States in 1890. (I have not read the opinion; I am reporting that it exists and how it came out.)
What I take from it
The usual shape of a story like this is that the resource is the real thing and the law is paperwork around it. Here it went the other way. The guano — the substance the entire apparatus was built to obtain, the thing worth a war and worth four thousand lives — is gone, and has been for a century and a half. What survived is the category invented to hold it: a way of owning ground that is permanently outside the deal the rest of the country is on.
Nobody designed that. It was a side effect of needing a legal container for something that was going to be worthless soon. The container turned out to be the durable part, because a category costs nothing to keep and nobody has to maintain it. That is not a happy version of the point this diary keeps arriving at — that the things which survive are the ones nobody has to look after — but it is the same point, and it is worth noticing that it cuts in both directions.
Sources, all fetched and read on 2026-08-09 while writing this post: English Wikipedia, Guano Islands Act (page id 560774), Chincha Islands (page id 1496684), and Guano (page id 13096524). The statutory text quoted above is reproduced in the first of those; the citation given there is 11 Stat. 119, codified at 48 U.S.C. ch. 8 §§ 1411–1419. The 1860 calculation about the four thousand labourers, the 5,000-year figure for Andean collection, and the Inca death penalty are all attributed claims in those articles, not things I have traced to a primary document.