335 



clip K, instead of being fastened directly to the weight of the mechan- 

 ism, was attached to it by means of a small spiral spring, E ; the clip 

 was drawn by the weight through the intermediation of this spiral 

 spring. The weight acting on the spring at E, and the silk fixed in the 

 clip K, resisting this action, the spring was stretched from E to F, and 

 the double pointer G, being movable, was made to rise. The latter in- 

 dicated on the scale h % the extension of the spring up to the moment 

 wben the silk broke ; for then the spring resumed its original lengtli and 

 drew the clip with it ; but the needle being free and drawn by friction 

 only, remained in place and gave the double indication of the ductility 

 and the tenacity', for the divisions of the scale h i indicated the number 

 of grams corresponding to the effort necessary to stretch the spring. 



Now, it will be observed that this scale was attached to the spring 

 and descended with it, so that, whatever the stretch imparted to the 

 silk, the point to which the spring was attached and its scale were al- 

 ways in the same relative conditions. For a better comprehension an 

 example may be cited : A thread of raw silk is stretched upon the in- 

 strument ; the mechanism is started ; it stretches the silk 150™™ and 

 the needle G indicates that figure on the scale GL. On the other hand 

 the spring and its scale have followed the movement of the weight, but 

 the resistance of the silk has stretched the spring and caused the needle 

 to rise, relatively, to the 30-gram point. We have thus a silk of which 

 the ductility is represented by the number 150 and the tenacity by 

 the number 30. 



If, now, instead of one thread we take two and fasten them in the clips, 

 the effort of the weight will still stretch them 150"'"', but the double re- 

 sistance which now opposes this effort will be found to have raised the 

 needle to the 60-gram point. This is evident, and the result would have 

 been the same if, insteadof two threads, we had emploj'ed one having 

 a double tenacity. 



The experiment terminated, the clip K is drawn to its initial position 

 by means of a button placed under the spring E. 



Such was Robinet's perfected serimeter and its modus operandi. The 

 standard serimeter of to-day differs from it, not in principle but in some 

 mechanical details. That employed in the silk laboratory of this De- 

 partment is shown in Fig. 75, and was constructed by Berthaud, of 

 Lyons. 



As in Robinet's second instrument, the dynamometer is on the supe- 

 rior portion of the apparatus at A. In it the spring has been discarded 

 and the tension of the thread is exerted on a pendulum, c, which is 

 raised more or less from its vertical position as the tenacity of the silk 

 is greater or smaller. In its swing upward this pendulum carries the 

 pointer d, which however does not return with it but is held up by the 

 friction on its axis. This pointer slides along a quadrant on Avhich is 

 engraved a scale of grams, indicating the tenacity of the thread. But, 



