Because of the destruction of most of the Maya codices in the sixteenth century, scholars had extremely limited access to the original texts. It was not until 1810 that the first reproduction of any Mayan codex— five pages from the Dresden Codex— were reproduced by Alexander von Humboldt in his Vues de cordillères, et monuments des peuples indigènes de l'Amérique. From this very limited reproduction in 1832 European-American autodidact polymath, mathematician, botanist, zoologist, and malachologist Constantine Samuel Rafinesque, while working in Philadelphia, deciphered the Maya's system of numerals.
In 1832 Rafinesque published his discovery in his periodical, the Atlantic Journal, and Friend of Knowledge: A Cyclopedic Journal and Review of Universal Science and Knowledge: Historical, Natural, and Medical Arts and Sciences: Industry, Agriculture, Education, and Every Useful Information. He announced it in a three-part article addressed to Jean-François Champollion, whose name he misspelled, "on the Graphic systems of America, and the Glyphs of Otolum or Palenque, in Central America." In the second part of this article, on page 42, Rafinesque briefly explained his discovery of the meaning of the Maya bar and dot system in which a dot equals one and a bar equals five.
"Later findings proved him right and also revealed that the Maya even had a symbol for zero, which appeared on Mesoamerican carvings as early as 36 B.C. (Zero didn't appear in Western Europe until the 12th century)" (http://www.pbs.org/wgbh/nova/mayacode/time-flash.html, accessed 10-10-2009).
Like most of Rafinesque's numerous other publications, his Atlantic Journal enjoyed very limited success, and folded after only eight issues. Copies of the original edition are extremely rare. My copy is a facsimile reprint issued by the Arnold Arboretum, Boston, in 1946.