310 ANKL-AL OF S< ILMliiC DISCOVERY. 



To test the drying rate of an oil, or of an oil mixed with some material 

 to dry it, it is simply necessary to spread it upon the surface of glass, and 

 expose it to the atmosphere, and to note the time occupied in its passing 

 from its fluid to a solid state ; and the circumstance that on its being 

 touched with the finger, and neither adhering to it on removal, nor resisting 

 its removal, is taken as evidence of its being dry. The drying rate of an 

 oil is materially affected by temperature the hygromctric condition of the 

 atmosphere, by its state as to stillness or motion, and by the presence or ab- 

 sence of sunlight, etc. An oil that in a warm, breezy summer's day will 

 dry in eight or ten hours, may not dry in less than sixteen or twenty when 

 the air is foggy and motionless ; and in an unfavorable winter's day may not 

 dry in less than from twenty-four to thirty hours. 



The practical results of a vast number of experiments made by the writer 

 are thus given : 



When an oil or an oil paint, on exposure to air, dries, four distinct kinds 

 of action come into play, to effect, or contribute to that result. These arc 



Firstly. The chemical actions taking place naturally between the oil and 

 the atmosphere, or the natural chemical action consequent on exposure. 



Secondly. Those due to some specific chemical action upon the oil of 

 some clement in the pigment, or the induced chemical action, as contra-dis- 



C5 



tinguisbed from the natural. 

 fi 1 



Thirdly. Those due to the peculiar physical structure of the composition 

 of the paint, through which, within the same superficial area, a larger surface 

 of the oil is brought under the action of atmospheric agencies. 



Fourthly. Purely mechanical actions, brought about by molecular dis- 

 turbances in the oil or paint, by which fresh particles of the oil or fresh sur- 

 faces of it arc being, during the time of exposure, continually thrown up to 

 the action of the atmosphere. 



In any instance of an oil or a paint drying, the influences at play to con- 

 tribute to that result can be traced to one or other, but most frequently to the 

 combined effect of two, or of all of those kinds of actions. 



The experiments showed that the hydrated protoxides of certain metals 

 pre-eminently exercise a specific drying action on the oil. The hydrated 

 protoxides of iron, of nickel, of cobalt, and of manganese, are the most re- 

 markable of the class ; but it is through the latter (i. e., the hydrated protox- 

 ide of manganese) that all the singular and happy effects upon the oil arc to 

 be practically accomplished. The addition to the oil of only from three to 

 five parts of the hydrated protoxide of manganese to 1,000 of the oil, gives 

 birth to peculiar changes ; and, by the simplest of methods of after treat- 

 ment, this chemical fact is made applicable to the production of any required 

 kind of drying oil. 



Arrangements for manufacturing drying oils on this principle have been 

 adopted, under the writer's direction. 



The oil, used in very large 'quantity at a time, is mixed with hydrated 

 protoxide of manganese, or materials yielding that, in quantity ranging be- 

 tween five pounds and fourteen pounds to the ton, and the oil warmed up 

 to from 100 to 150 deg. Fah. In a very short time ten or twenty minutes 



