174 LETTER TO GORDOX. 



I have communicated a plan to Newall how this is very 

 easily to be done. Then faulty weldings will be disclosed by 

 rupture, and one may feel pretty safe afterwards. Furthermore 

 a dynamometer of solid iron must be constructed with an 

 accurate scale, and in such a way that with the maximum 

 loading there still remains a deflection of at least one foot. It 

 is better to make use of a well-made spring than a weight, so 

 that the fluctuations of the apparatus may be as small as pos- 

 sible. It would also be very advisable to carry the wire behind 

 the brake over two fixed and one moveable pulley, the latter 

 being pulled doAvii by a weight or still better by a very 

 strong spiral spring. The up and down movements of the 

 ship are thereby rendered innocuous. 



28th. 



As Loffler has not yet returned, I can communicate to 

 you nothing definite in regard to the calculated forces. You 

 are quite right that the assumed forces are not justified by the 

 depths alone. I believe we may go to half the depth for which 

 a cable can support itself, with tolerable safety and to a third 

 with great safety. Up to a fifth of the depth 5 to 10%, to a third 

 10 to 15% slack may give sufficient safety, if the weather is 

 favourable. At greater depths the loss must be considerably 

 more. Newall's plan of retarding the sinking of the cable 

 by shields is wrong in principle. The cable must sink as 

 quickly as possible on account of the currents. With moderate 

 depths it is more advantageous to take back the slack by 

 somewhat greater loading of the brake. If the depth is greater 

 than y 3 to l / 2 of the minimum strength of the cable, the sliding 

 back of the cable must be slackened as far as possible by 

 disks attached at right angles to the cable. I believe these 

 are best made of sheet-iron. A few large ones are far more 

 effective than many small ones. The attachment can be effect- 

 ed in many ways. One must then proceed as quickly as 

 possible, in order to keep the angle acute. For the measure- 

 ment of velocity I am now having made an electrical apparatus, 

 which turns a large indicator by the side of the brake. The 

 brake-wheel must indicate in the same way, so that at any 

 moment the proportion of the velocity and of the exerted force 



