But 1 gallon of dry alcohol, extended in 2 gallons of water, gives 3 gallons of liquor at 19 deg., which is called Holland, or first proof; a produce surpassing all what has been hitherto known to the distillers. I will prove it by an example: 1 gallon of molasses yields only 1 gallon of rum, at 19 deg., to the rum distiller; still, molasses is a true sirup, composed of 8lbs. of sugar, or sweet matter, more fermentable than sugar. 12-1/2 gallons of molasses, representing 100lbs. of dry sweet matter yield consequently 12-1/2 galls. of rum, Holland proof, which is only half the produce obtained by Lavoisier; an immense difference capable of exciting the emulation of all distillers, as it proves the imperfection of the art. What are the causes of such a dissimilarity of product? We must seek for them. 1st. In the difference of the strength of the vinous liquor. Lavoisier employed only 4 parts of water to 1 part of dry sugar. The rum distiller usually puts 10 gallons of molasses to 90 gallons of water, or the residue of the preceding distillations. 10 galls. molasses contain 80 lbs. of sweet matter. 90 gallons of water weigh 720lbs.; therefore the proportion is, one part of sweet matter to 9 parts of water--whilst that indicated by Lavoisier is only 4 parts of water to 1 part of sugar.[A] It is obvious how much richer this last must be, and that the fermentation thus produced has an energy far superior to the other. Thence results a rapid production of spirit, operated in a short time; whilst that of the rum distiller languishes more or less, and a slow fermentation wastes part of the spirit which it produces, even as it is forming. 2dly. Bodies evaporate in proportion to the extent of their surface. One hogshead of 100 gallons, should contain, according to Lavoisier's composition, the elements of 50 gallons of spirit, at 19 deg.; whilst that of the rum distiller contains only 12. Now, as every fermentable liquor requires open vessels, the hogshead of the rum distiller loses as much spirit as that of Lavoisier: hence it is plain how far the above proportion operates to the disadvantage of the fermer.