ness. For the portholes of the bathyscaphe FNRS 2 I chose an interior 

 diameter which was only two-thirds of the thickness. Cross No. i 

 is the record of one of these portholes under a load of 2-5 miles of 

 water. It may be concluded from these tests, and from a glance at our 

 diagram, that the portholes of the Trieste and of the FNRS 3 offer 

 complete safety at the depths for which our bathyscaphes have been 

 designed. It would be interesting to know the pressure at which our 

 portholes would be driven in. As long as tests have not been made for 

 pressures much greater than 1600 atmospheres, an exact answer cannot 

 be given to this question. By roughly extrapolating the results shown 

 by our diagram I think that it would be only at depths exceeding 12 

 miles that our plexiglas would show a permanent deformation. 

 Breakage would only occur at much greater depths, and there is no 

 part of the ocean depths known to exceed about 7 miles. 



It goes without saying that our tests with models are only conclusive 

 upon condition that the material of the real portholes has the same 

 mechanical properties as that of the models. Now the models were cast 

 in one piece. For the fabrication of the portholes it was necessary, in 

 order to avoid excessive heating at the moment of the polymerization 

 of the methacrylate, successively to cast several layers of the raw 

 material one upon the other. We made sure that this procedure was 

 equivalent to that of casting a single piece by cutting small test pieces 

 out of the material removed during the machining of the large port- 

 holes by the Vetrocoke Company. The tensile tests effected, for 

 example, with test pieces whose direction of loading was at an angle of 

 45° with the layers, have shown that the bonding between these layers 

 was perfect. 



To conclude this survey, let us mention one more interesting fact : 

 if a piece of plexiglas is overloaded, its whole mass becomes cloudy; 

 if the overload has not been too great, this cloudiness will disappear 

 with the load. If, on the contrary, the piece has been greatly over- 

 loaded, the cloudiness will persist, even after the load has been released. 

 It is unnecessary to say that in our real portholes we have never 

 observed the formation of the faintest cloudiness. This is one more 

 reason for our confidence. 



[151 ] 



