Although Evangelista Torricelli is universally credited with inventing the barometer in 1643, historical documentation also suggests Gasparo Berti, an Italian mathematician and astronomer, unintentionally built a water barometer sometime between 1640 and 1643. French scientist and philosopher René Descartes described the design of an experiment to determine atmospheric pressure as early as 1631, but there is no evidence that he built a working barometer at that time.
On July 27, 1630, Giovanni Battista Baliani wrote a letter to Galileo Galilei explaining an experiment he had made in which a siphon, led over a hill about twenty-one meters high, failed to work. Galileo responded with an explanation of the phenomenon: he proposed that it was the power of a vacuum that held the water up, and at a certain height the amount of water simply became too much and the force could not hold any more, like a cord that can support only so much weight. This was a restatement of the theory of horror vacui (“nature abhors a vacuum”), which dates to Aristotle, and which Galileo restated as resistenza del vacuo.
Galileo’s ideas reached Rome in December 1638 in his Discorsi. Raffaele Magiotti and Gasparo Berti were excited by these ideas, and decided to seek a better way to attempt to produce a vacuum than with a siphon. Magiotti devised such an experiment, and sometime between 1639 and 1641, Berti (with Magiotti, Athanasius Kircher and Niccolò Zucchi present) carried it out.
Four accounts of Berti’s experiment exist, but a simple model of his experiment consisted of filling with water a long tube that had both ends plugged, then standing the tube in a basin already full of water. The bottom end of the tube was opened, and water that had been inside of it poured out into the basin. However, only part of the water in the tube flowed out, and the level of the water inside the tube stayed at an exact level, which happened to be 10.3 m, the same height Baliani and Galileo had observed that was limited by the siphon. What was most important about this experiment was that the lowering water had left a space above it in the tube which had had no intermediate contact with air to fill it up. This seemed to suggest the possibility of a vacuum existing in the space above the water.
More info here: http://www.laamash.com/