French Wikisource, 2026-10-06 (Investigator, Step I, run 1)
Site: https://fr.wikisource.org (reached through Chrome; no blocks).
Query that found it: Google Poincaré "La mécanique nouvelle" Lille 1909 Mercure 32".
La Mécanique nouvelle (Lille), Poincaré's own longer twin of the project essay (priority 1)
URL: https://fr.wikisource.org/wiki/La_M%C3%A9canique_nouvelle_(Lille) Edition shown: Gauthier-Villars, 1923, pp. 1-17, preface by Édouard Guillaume. Embedded page markers [171]-[177] fit the Revue scientifique printing that Walter (2008) cites: "Revue scientifique, 12, p. 170-177" (see google notes) [web]. Setting [web]: Association française pour l'Avancement des Sciences, Lille congress, 1909. Lecture given on Tuesday 3 August at 8:30 pm in the Grand Théâtre, after Poincaré received the Association's grande Médaille d'or from its president, Professor Landouzy.
The Lille text uses much of the Göttingen text word for word (radium artillery, Paris/Berlin observers, 1/200 000 000, gaps in the ether, the bird mistaken for Vulcan, the pedagogic danger). It also contains material that the 1913 Monist translation does not have. Each item below is quoted or closely paraphrased from the French [web]:
- Mercury, the 6" figure (claim 5). "la mécanique nouvelle rend bien compte du sens de l'erreur relative à Mercure, mais elle nous donne seulement un mouvement de 6″ ; elle laisse donc encore une marge de 32″". So 38 - 6 = 32 is stated outright here, where the English Monist text only implies it (deep read: "38 - 32 = 6 is an inference, not printed"). The excess is described as "Sa vitesse angulaire est de 38″ plus grande qu'elle ne devrait être". Like the English, the French gives no time period. Lille adds that the perihelion is "le seul élément qui soit atteint" and that the theory of the Moon is not affected either.
- Experiments named (claim 1). The Göttingen/Monist text names no experimenter for the radium deflection. Lille names them: "Les expériences les plus récentes sont celles de M. Bucherer ... ce n'est pas celui auquel conduisent les expériences de MM. Kaufmann et Bucherer". It explains the method: beta rays (not alpha or gamma), a Faraday cylinder showing their charge, and comparing the electric and magnetic deflections to get both the speed and the inertia-to-charge ratio. It gives the speeds as a tenth to a third of light speed (30,000 or 100,000 km/s).
- Direction-dependent mass (claim 4). Lille explains why inertia differs for a force along the velocity and for one across it ("si la vitesse est grande, cette résistance ne serait pas la même dans les deux cas"). These are the longitudinal and transverse masses. The essay has only one clause on this [read: p.8].
- Self-induction as the "inertia of the ether" (claim 4). "Cette inertie de l'éther est un phénomène bien connu ; c'est ce que les électriciens appellent la self-induction". The example is the spark when a trolley pole leaves the wire.
- Aberration test of absolute motion (claims 2-3). Figures 26-27 show stellar aberration for two diametrically opposite stars. Comparing their annual amplitudes would reveal the Sun's absolute speed. Poincaré says he is "absolument sûr d'avance" that no difference will be found, because the astronomers' graduated circles are themselves deformed in a way that gives "une parfaite compensation". He also explains the light-wave geometry: spherical wave surfaces look ellipsoidal to a contracted observer, so the apparent light-time from A to B is "le même que si la translation n'existait pas".
- Inertial mass versus gravitational mass (claim 5). "Nous devons donc distinguer la masse, coefficient d'inertie, et la masse, coefficient d'attraction". Whether the attraction coefficient grows with velocity "est là une question que nous n'avons aucun moyen de décider". This is the same question that later became the equivalence principle.
- Lorentz's gravity as unequal electric attraction (claim 5). Lorentz's hypothesis (in Poincaré's account) is that a +1 charge attracts a -1 charge slightly more than two like charges repel. Gravity is then a leftover electrical force: "C'était, en somme, l'hypothèse de Franklin". It therefore obeys electrodynamics and the principle of relativity.
- Laplace's speed of gravity (claim 5). Laplace's result that gravity travels "au moins dix millions de fois plus vite que la lumière" assumed an aberration-like force direction. Poincaré calls this "une hypothèse toute particulière, assez mal justifiée ... Le résultat de Laplace ne prouve rien contre la théorie de Lorentz."
- Planetary radiation sized (claim 5). The acceleration of celestial bodies is "quelque chose comme un milliard de fois plus petite qu'à l'anticathode d'un tube de Crookes". The radiation would be imperceptible, but it exists.
- Galileo remark. Lille adds: "je n'ai jamais eu une telle pensée ; c'est bien pour la vérité que Galilée combattait".
Bearing: this is the inventor's own fuller statement of the same lecture, given three months after Göttingen. It settles the 6" reading of claim 5. It names the experiments behind claim 1. It shows that Poincaré treated the inertial/gravitational mass question and Laplace's objection as open questions. Open question for the reviewers: is the missing 6" and Bucherer material in the Monist text Halsted's cut, or was it already absent from the Göttingen French (Sechs Vorträge, 1910, pp. 49-58)? That volume was not checked this run.