252 ANNUAL OP SCIENTIFIC DISCOVERY. 



rnent was tried, the anxiously-looked-for experiment proved a failure. 

 Instead of a loud explosion, and the lumps of clay blown in fragments into 

 the air, a noise that could hardly be called an explosion was heard, and 

 the weights were found to be undisturbed. The experiment was again 

 tried, and again with results little better. The spout was supposed to be 

 too nearly on a level ; it was brought out of the drift- way and led up the 

 hill, so as to be nearly perpendicular ; and when the water this time also 

 increased in quantity was poured in this manner, a greater explosion 

 was produced, and considerable disturbance of the superabundant materials 

 was manifest ; but still the effect was not equal to what had been antici- 

 pated. It was then determined to remove the clay altogether, and to try 

 the effect of placing one iron pot inverted over the top of the other. Ey 

 this time the repeated failures had emboldened the spectators, and several 

 of them were now induced to join Captain Du Cane, who from the begin- 

 ning had placed himself in an inconvenient proximity to the scene of the 

 experiment, which nobody else had, up to this time, cared to imitate. 

 The materials were again ignited, the word was given, the water was poured, 

 and a miserably disappointing " puff" was again the result. The spectators 

 looked blankly at each other, and then a step or two was made to examine 

 the cause of the failure ; when, crash ! a tremendous explosion was heard, 

 and the iron pot which rested on the top of the burning mass was sent 

 soaring high xip into the air, a height of from 50 to 70 feet. A few mo- 

 ments longer, and it would have encountered in its first progress the heads 

 of some half dozen curious gazers in the drift- way. 



ON THE MEANS TO BE EMPLOYED FOR DETECTING AND REN- 

 DERING PERCEPTIBLE FRAUDULENT ALTERATIONS IN PUBLIC 



AND PRIVATE DOCUMENTS. Chevallier and Lassiagne. 



The numerous experiments which have been already tried at various 

 times, have made known the processes which may frequently be put in 

 practice for causing the reappearance of traces of writing effaced by chemi- 

 cal reactions, and for throwing light on the work of the guilty. But there 

 are cases in which all the means proposed for this purpose fail, and then 

 the criminal may escape justice from the want of conclusive material proofs. 

 If, as has already been proved, it is not always possible to cause the 

 reappearance of the effaced writing, for which other written words have 

 with a fraudulent intent been substituted, at least, as our experiments 

 demonstrate, we may recognize, by some effects which are manifest on the 

 surface of the altered paper, the places where the criminal act has been 

 performed, circumscribe them by a simple chemical reaction visible to the 

 least practised eye, and even measure their extent. In a word, the invisi- 

 ble alterations produced on a deed are susceptible, owing to the partial 

 modifications which the surface of the paper has undergone, of being 

 differently affected by certain chemical actions, and of being rendered 



