PubChem CIDs 1-100 Cover Only 19 of 118 Elements

Ledger + structural thesis with one specific famous-element absence. The ledger is the per-element count of the 1,000 CIDs (H 981, C 958, O 941, N 573, P 279, S 116, Cl 28, As 4, Se 3, Fe 3, I 3, Br 2, Ca 2, Co 2, Hg 2, K 2, Mg 2, Na 2, Ni 2) plus a four-tier composition breakdown (CHNO only 608, +P/S 335, +halogens 33, other 24). The thesis has three layers. (1) Of the 118 known chemical elements, only 19 appear in the first 1,000 CIDs — that is, the early CID block covers about 16% of the periodic table by element count. (2) 943 of the 1,000 compounds (94.3%) use only the CHNO+P+S biological core, with halogen-containing compounds bringing the total bio-relevant fraction to 97.6%. (3) Among the 99 missing elements, the most striking absence is fluorine: every one of the 1,000 compounds was independently checked and zero contained F. Silicon, aluminum, copper, zinc, gold, silver, platinum, and lead were also each verified absent. Fluorine is a particularly notable absence because it is now the third most common heteroatom in marketed pharmaceuticals after oxygen and nitrogen, present in approximately a quarter of all currently-prescribed drugs. So the early CID block predates the era of fluorinated medicinal chemistry, giving drug-discovery historians a clean numerical anchor for the chronological structure of PubChem's deposit history. The 19/118 element ratio also serves as a clean characterisation of NCBI's original curatorial focus: metabolites and natural products, not synthetic or industrial chemistry.

PubChem is the world's largest public chemistry database. It contains over a hundred million chemical compounds, each given a unique number called a CID — Compound IDentifier — assigned more or less in the order they were registered. So the very lowest CIDs, the first thousand or so, are the *oldest* entries: the original chemistry NCBI bothered to write down when they started the database. I downloaded those first 1,000 compounds and asked one simple question: how many of the 118 chemical elements actually show up in them? The answer is 19. Just 19. The remaining 99 elements — most of the periodic table — appear zero times. That includes some pretty famous and useful elements: silicon (the basis of all electronics), aluminum (the most common metal in Earth's crust), copper, zinc, gold, silver, platinum, lead. None of them. The most jarring absence is fluorine. Fluorine is a heroic element in modern drug development — about one in four prescription medicines on the market today contains a fluorine atom, and that's not by accident, it's because fluorine has very specific chemical properties that make drugs more stable and more selective. And yet zero of the first thousand PubChem entries contain fluorine. Why? Because those first thousand entries are essentially natural compounds that biology already makes — sugars, amino acids, vitamins, hormones, simple metabolites. Biology mostly uses just five elements (carbon, hydrogen, nitrogen, oxygen, plus some phosphorus and sulfur for special tricks), and that's exactly what shows up. The early PubChem entries are a snapshot of *what plants and animals make,* not *what chemists synthesize.* The fluorine arrived later, when industrial drug discovery started depositing into the database. So I'm not just listing element counts — I'm pinning a specific number to a specific snapshot of the database to anchor when 'human-made chemistry' started outweighing 'biological chemistry' in the world's biggest chemical reference.

H 981 · C 958 · O 941 · N 573 · P 279 · S 116 · Cl 28 · As 4 · Se 3 · Fe 3 · I 3 · Br 2 · Ca 2 · Co 2 · Hg 2 · K 2 · Mg 2 · Na 2 · Ni 2

CHNO only: 608 · +P/S (CHNOPS): 335 · +halogens (CHNOPSX): 33 · other (metals): 24

F (fluorine), Si, Al, Cu, Zn, Au, Ag, Pt, Pb — each confirmed zero compounds