SCIENTIFIC SUMMARY. 
313 
gathers from Thorpe’s paper that “the results are such as apparently esta- 
blish its great exactness, while in simplicity and ease of execution it would 
seem to he quite superior to the similar methods which have been previously 
proposed.” The first determination which Johnson made by Thorpe’s pro- 
cess showed a deficiency of 4o per cent. In a second experiment nitric 
oxide was evolved, “ the standard acid not only not being neutralized by 
ammonia, coming from reduction, but made more acid by the reaction of 
nitric oxide upon the oxygen and water of the condensing vessels.” In the 
third and fourth experiments the results were equally unsatisfactory. The 
author then made three determinations by Bunsen’s method with zinc-iron 
couples and caustic potash, and obtained concordant and perfectly satisfac- 
tory results, the numbers, it should be stated, being uniformly 0-3 to 0*6 per 
cent, under that required by theory . — American Journal of Science , April, 
1877, xiii., 260. 
The Protection of Iron against Rust . — A most important method has been 
devised by Professor Barff for preserving iron from rust. As far as we are 
able to judge from the results of the experiments which have thus far been 
made, the process is one of manifold application, and bids fair to prove of 
the greatest service in many branches of industry. Professor Barff has dis- 
covered a means of rendering all kinds of ironwork, however much exposed 
to the weather or to corrosive vapours and liquids, practically indestructible 
and permanent. Iron when exposed to the action of water or moist 
air begins to rust, a film of ferrous oxide being in the first instance 
formed upon its surface ; this rapidly takes up more oxygen from the air, 
and a higher oxide, the sesquioxide, is formed ; the latter compound gives 
up some of the oxygen to the unchanged metal beneath it, and the fresh 
ferrous oxide thus produced slowly unites with more oxygen, which 
traverses the porous layer of sesquioxide overlying it ; in this manner the 
change is propagated to greater and greater depths, until in process of time 
the whole of the metal may be converted into rust. Various methods are 
employed to check this oxidation ; paints and varnishes are used with, how- 
ever, only partial success, from the fact that the adhesion of those materials 
to the metallic surface is imperfect, and they are liable to scale off aud dis- 
integrate with changes of temperature. Professor Barff proposes as a remedy 
the covering of the face of the metal with a layer of the oxide of iron, inter- 
mediate in composition between the two compounds we have above alluded 
to, the ferroso-ferric or magnetic oxide ; and this he accomplishes by ex- 
posing the metal to the action of superheated steam at a high temperature. 
Iron treated in this manner for from six to seven hours at 1200° F. becomes 
covered with a black film of magnetic oxide, which adheres to it even more 
firmly than the metallic particles adhere to each other, and is sufficiently 
hard to resist the action of a file. At his lecture delivered before the Society 
of Arts, and subsequently at a soiree held in the Royal Society’s rooms, 
Professor Barff showed specimens, treated by his method, which had passed 
unscathed through a six weeks’ exposure to bad weather on a lawn in Bays- 
water, as well as others which had been lying in contact with corrosive liquids 
of every kind in the sink of a laboratory. It is easy to conceive of applications 
being made of Professor Barff s method which may prove of the greatest 
value and importance. Among many which have been suggested are the 
NEW SERIES, YOL. I. — NO. III. Y 
