Attempts to differentiate between wrought iron made by three different processes. The bloomery process, which involves direct reduction from the ore to low carbon iron, was the first iron production process in use, but fell into disuse (in Europe and North America) by the late 19th century. The finery process, in which molten cast iron was decarburized, developed in the late 16th century in Europe, only after cast or pig iron was widely available. The puddling process, developed by the late 18th century in Britain, superceded the other two. The slag associated with production is trapped in the metal and can be used to differentiate the production methods. The puddled wrought irons have the purest slags, with upper limits of 24% silica, 90% wustite (FeO), and 1% alkali. Bloomery irons have much less pure slags, with 48% silica, 70% combined iron oxide and Fe₂O₃, and 4.6% K₂O. Finery slags seem to be intermediate, but there are fewer published or reliable analyses on which to base conclusions. An iron rod and a steel strap from 19th-century Jamaica were examined, and it was concluded that the rod was not made from puddled iron, but it could not be decided (on the basis of the slag composition) whether it was made by the bloomery or finery process. The steel strap is likely to be of a bloom iron origin. As there are many sources of variation in the analysis of slag inclusions, statistics were based on the Weibull distribution, rejecting the Normal (Gaussian) or log normal distributions as not fitting the data well. It is also demonstrated that at least 10 replicate analyses of large inclusions (ca. 450 micrometers long by 70 micrometers wide) should be analyzed by energy dispersive spectrometry (EDS) before conclusions can be drawn.