The exposition in the web page [https://www.bestor.be/wiki/index.php/Voyager_dans_le_temps._L%E2%80%99introd...]      gives several contemporary sources from which one can conclude      that the switch in Europe/Brussels on 1892-05-01 was      from 00:17:30 to 00:00:00, and not from 12:00:00 to 11:42:30      as is currently contained in tzdb.      It is true that mean time at the Royal Observatory was      determined to be UT + 17 min + 29 s at the time, and that      the observatory delivered Brussels time every morning at 06:00,      but this does not mean that Brussels time was UT + 17 min + 29 s      because the observatory had recently moved to Uccle (see      [Ernest Pasquier: "L’unification des heures", page 306]      as referenced in the web page above).      The text in tzdb            LMT before 1892 was 0:17:30, according to the official            journal of Belgium: Moniteur Belge, Samedi 30 Avril 1892, N.121.      is probably incorrect as regards the "according to".      The text of the 1892 law is reprinted in         [Laurent Delatte: "C'était dans Ciel et Terre il y a cent ans".          Ciel et Terre vol 108 p 113..116. 1992.], online at [http://articles.adsabs.harvard.edu/cgi-bin/article_queryform?bibcode=1992C%2...]      on page 115, and it does not contain any offsets from UT.      And it hardly could contain offsets because every larger town      in Belgium had its own railway time, and civil time outside the      stations often differed, as described by the articles by      Ernest Pasquier referenced in the web page.      Anyway, the actual time step taken was 17 min + 30 s, and time      jumped from 00:17:30 to 00:00:00. This is described in the circular      issued by the administration of railways and quoted in      [Gazette de Charleroi of 1892-04-30, "Réglez vos pendules" page 1..2],      and linked to in the BEstor web page.      Michael Deckers.