312 PROCEEDINGS OF SECTION G. 



Experiment No. 8. — 453 grammes of bright 2-inch wire 

 nails were exposed in a photographic dish on a window sill 

 from September 11, 1891, to February 16, 1892. These 

 nails were moistened from time to time, but not kept con- 

 stantly wet. On scraping off the rust with a horn spatula, 

 7-79 grammes of rust were obtained, the whole of which was 

 magnetic. 



Experiment No. 9. — 453 grammes of 6-inch bright wire 

 nails were similarly exposed and treated at the same time. 

 These nails only yielded 4'03 grammes of rust, which was 

 also wholly magnetic. The smaller surface exposed by the 

 larger nails accounts for the smaller weight of rust produced. 



Experiment No. 10.— On September 21, 1891, 906 

 grammes of bright 6-inch wire nails were put out on a 

 window sill (facing W.) in a new photographic dish, and 

 covered with distilled water. After three days, i.e. on the 

 24th, they were covered with loose ochrey rust which washed 

 off readily, but under this and fUhng the striae of the nails 

 was a film of black oxide. When scraped on February 1 6, 

 1892, with a horn spatula, they yielded 4*03 grammes of rust 

 of a dark brown colour, and entirely attracted by the magnet. 



Experiment No. 11. — At the same time and place l^lbs. 

 of bright 2^-inch wire nails were also put out in a photo- 

 graphic dish with the same result, except that there was very 

 much more of the black oxide present. 



On turning over the nails Nos. 10 and 1 1 on the 26tli, i.e. 

 after five days' exposure, it was found that, while covered 

 more or less completely with ochrey rust on the upper 

 surface, the lower surfaces of the nails (which were com- 

 pletely immersed in the water) were free from it, but coated 

 instead with a closely adherent film of black oxide, due 

 apparently to the fact that, while there was sufficient oxygen 

 on the exposed side of the nails to form both the red or 

 ochrey and black oxide on the lower side (completely 

 immersed in water) the oxidation did not go beyond the first 

 stage. On February 16, 7*79 grammes of entirely magnetic 

 rust were obtained. 



The total amount of rust furnished by this experiment and 

 No. 10 was much more than the weights given. The loose 

 rust in the dishes was not weighed, because it was mixed with 

 grit and dust which had fallen into the dishes during their 

 exposure. 



Experiment No. 12. — Some sheets of clean iron were also 

 placed out but not moistened. The rust was much darker 



