28 



Mr. J. R. Green. 



[Jan. 14, 



hollow circular cylinder of the same thickness and the same sectional 

 area. Experiments npon the torsional strength of hollow prisms of 

 various forms, having the same sectional area and thickness of shell, 

 can alone determine the latter point ; while, at the same time, such 

 experiments would serve the further purpose of showing how the 

 condition above referred to — that the shell shall be stiffened internally 

 so as to effectually resist change of form — can best be complied 

 with. 



The distribution of the torsional stresses over the transverse section 

 of a ship's hull is obviously different from the distribution of the 

 stresses due to longitudinal bending. The parts subjected to greatest 

 stress by twisting are those which are near to the centre of gravity of 

 the transverse section ; and they are the side plating near the neutral 

 axis of longitudinal bending in the upright position and the middle 

 portions of the plating of the decks. Those parts of the hull which 

 are usually made the strongest, viz., the strakes of side and bottom 

 plating that are farthest from the neutral axis, and the upper deck 

 stringer plate, are those which are least affected by twisting. It is 

 probably owing, in great measure, to the straining action caused by 

 twisting, that experience has proved it to be necessary to make the out- 

 side plating of a ship of nearly uniform thickness over the whole section; 

 and it cannot be because of the reason sometimes given, that the 

 plating in the vicinity of the neutral axis when a ship is upright is 

 often brought by rolling into positions in which it is greatly strained 

 by longitudinal bending. 



The importance of many of the structural arrangements of ships 

 that are described in the present paper, which practical experience 

 has shown to be necessary, may be understood from these considera- 

 tions ; and it will also be seen that no rules for regulating the strength 

 of ships are likely to be satisfactory if based, as is often done, upon 

 the hypothesis that the straining actions caused by longitudinal 

 bending are so much more important than all others that it is suffi- 

 cient to regard them alone. 



III. " Proteid Substances in Latex." By J. R. Green, B.Sc., 



B. A., Demonstrator of Physiology in the University of 

 Cambridge. Communicated by W. T. Thiselton Dyer ? 



C. M.G., F.R.S., Director of Royal Gardens, Kew. Received 

 January 4, 1886. 



In the study of the metabolism of plants, the nature of the pro- 

 ducts resulting therefrom, and the different forms assumed by these 

 bodies during the changes involved, attention has been chiefly 



