168 ANNUAL RECORD OF SCIENCE AND INDUSTRY. 



siderably less. The velocity at various depths, according to 

 Henry, may be represented by the ordinates of an ellipse, 

 whose abscissas represent the depths below the surface. In 

 order to obtain the velocity of any cross section of the stream, 

 Henry maintains that the mean velocity observed at six 

 tenths of the depth, especially when the observations are 

 made at a number of stations in a line crossing the stream, 

 is decidedly better than the mean velocity deduced from mid- 

 depth observations. 4 2>, 1873, 155. 



A NEW EXPLANATION OF EUPEET's DROPS. 



The bursting produced in Rupert's drojDS the moment that 

 the thin end is broken off has been hitherto attributed to a 

 state of forced dilatation of the interior. 



These drops, as is well known, are generally made by drop- 

 ping melted glass into a basin of cold water, and it is sup- 

 posed that the exterior surface being suddenly chilled, con- 

 lines the interior portions in a condition such that the least 

 disturbance of the equilibrium of the outer surface allows 

 the interior to suddenly expand, thereby causing the w^hole 

 piece of glass to fly into numerous fragments. De Luynes 

 has quite carefully investigated this subject of the break- 

 ing of the Rupert's drops, and, in order to cause them to 

 break without applying to them from the exterior any addi- 

 tional mechanical force, he suspends such a drop by a thread 

 over a platinum vessel containing fluohydric acid. On al- 

 lowing the fine extremity of the glass to dip into the liquid, 

 he finds that he can always dissolve the whole of the thin end 

 without destroying the drop. But when the acid touches the 

 origin of the neck, that is, its point of divergence, the equi- 

 librium, is always broken, and the drop then separates into 

 a great number of fragments, but in most instances without 

 explosion. On the other hand, if he dip the large end of the 

 drop into the acid, keeping the origin of the neck and the 

 whole of the thin end out of the liquid, the drop may be com- 

 pletely dissolved Avithout rupture. 



These experiments prove that the stability of the drop de- 

 pends on the existence of the origin of its neck. If, again, 

 we cut through the drop by a saw, the explosion always 

 takes place as soon as a little more than half the thickness 

 of the glass has been cut away. Again, if the drops are en- 



