batteries of the FNRS 2 was taken over for the construction of the 

 FNRS 3. (This is proved by the fact that the FNRS 3 lost its batteries 

 several times !) The tanks of the FNRS 2, filled as an economy measure 

 with gravel and scrap iron, were provided, at their base, with a hatch 

 which could be opened in the middle of a dive, but since they could 

 only be closed on the surface, they were simply replaced by a large 

 vessel filled with lead pellets, the bottom of the vessel operating 

 exactly as with my reservoir tanks. In spite of its inconveniences, the 

 principle of my mixed system without the 'blocks of cast-iron' (which, 

 for that matter, were boxes filled with scrap iron) was thus retained for 

 the FNRS 3. It was for the Trieste and not for the FNRS 3 that my 

 son Jacques suggested giving up the mixed system and employing 

 only magnetic pellets. It is a great simplification. The iron pellets can 

 thus be allowed to flow out in small doses, while, in case of emergency, 

 the two tubs containing this ballast can be released. 



EMPTY TESTS WITH OVERLOAD 



I must first of all explain to the uninitiated the reason for the empty 

 tests. 



Each piece in a construction must be calculated with a certain margin 

 of safety to resist the stress that it should normally bear. If there is no 

 error in the calculations and if the material itself is in exact con- 

 formity with the premises of the calculations, the piece will certainly 

 resist. Experience proves, however, that there can be defects in the 

 materials. This is particularly true of a piece of metal which is not 

 forged but cast (such as the cabin of the FNRS 2-3). To make sure of 

 the strength of the piece one makes an 'empty trial'. 



However, we know that if a part is loaded several times almost up 

 to its initial breaking limit it will become 'fatigued' and will end by 

 giving way even with loads somewhat inferior to the first. That is 

 why one must make an empty trial with a certain overload. In the 

 greater number of cases, an overload of the order of 50% is chosen, 

 this being often even imposed by legal regulations. It is for this reason 

 that, from the beginning of the building of the FNRS 2, 1 had declared 

 that it should never be used by men at depths exceeding two-thirds of 

 the trial pressure of the cabin, and I had provided for first sending it 

 down empty to 3 J miles, if we ourselves had to go down to 2^ miles. 

 This cabin is made of cast-steel. It contains, like many castings, small 

 air-holes, with diameters reaching y in. visible upon the radiographs, 



[ I8<S ] 



