March 13, 1903.] 



SCIENCE. 



417 



The crust of malachite which firmly 

 bound these pieces together varied in thick- 

 ness up to two millimeters, and within this 

 there was a layer of red oxide of copper of 

 similar thickness which is almost free of 

 silver, containing but about one per cent. 

 Inside of these two crusts there remained 

 more or less of a dark spongy mass of 

 silver, retaining a little oxide of copper 

 which adhered to the unchanged alloy, and 

 in some instances the latter had entirely 

 disappeared, so that this residue of the 

 coin was fragile while still partly retain- 

 ing the imprint of the coin. The manner 

 in which the silver was concentrated is of 

 decided interest. 



From the contents and circumstances of 

 this find these coins had no doubt been 

 thus stored away for nearly 1,900 years, 

 and yet on most of them the lettering and 

 even the dates may be deciphered. 



A Convenient Form of Table for Calcula- 

 tions of Chemical ^¥ eights: F. P. Dun- 

 NINGTON, University of Virginia. 

 The author, having frequent occasion to 

 check calculations of the amounts of sub- 

 stances from the weights made by students 

 in quantitative analysis, has constructed a 

 table to enable him quickly to obtain such 

 results. A description of this is given to 

 the section as a proposed accessory in teach- 

 ing, which may prove instructive to the 

 pupil and helpful to the teacher. 



Upon coordinate paper 500 by 400 mm. 

 the 500 mm. abscissa is counted as 1,000, 

 and there is laid off: upon the (vertical) 

 ordinate the length corresponding to the 

 figure expressing the amount of body 

 sought for each 1,000 parts of the sub- 

 stance found. A straight line is then 

 drawn from the origin to this point, as 

 798 for the copper in 1,000 of CuO, or 

 247 for the chlorine in silver chloride. 



Similarly a diagonal line is drawn from 

 the origin for each such form in which 



bodies are weighed. If, then, one reads 

 olf upon the abscissa the measure corre- 

 sponding to any weight of substance found, 

 the length of the ordinate at that point 

 which is cut off by the corresponding diag- 

 onal gives directly the amount of the body 

 sought. For instance, .679 of MnS gives 

 .429 of Mn. 



The Action of TJnsymmetrical Hydrazines 

 on the Chlorine Derivatives of the Qui- 

 nones of the Benzene Series: William 

 McPherson, W. L. Dubois and C. P. 



LiNVILLE. 



The unsymmetrical aeyl phenyl hydra- 

 zines react with the different chlor-qui- 

 nones in the following ways: (1) A regular 

 condensation may take place forming the 

 corresponding hydrazones, which on sapon- 

 ification yield oxyazo-compounds. (2) A 

 hydrogen atom of the quinone together 

 with a hydrogen atom of NHj group of 

 the hydrazine may be removed by the 

 oxidizing action of the quinone, forming 

 a hydrazino compound. (3) A chlorine 

 ■atom of the quinone may combine with a 

 hydrogen atom of the NHj group of the 

 hydrazine, a hydrazino compound being 

 formed with the evolution of hydrochloric 

 acid. 



Some New Phenomena Exhibited by Soap 



Solutions: H. W. Hillyee. 



As the formation of bubbles is depend- 

 ent on the low surface tension of soap 

 solution, so the emulsifying power of soap 

 solutions is largely dependent on the low 

 surface tension between the soap solution 

 and the oily matter removed by it. This 

 lowering of the surface tension is, within 

 certain limits, nearly proportional to the 

 amount of soap in solution. Neither of 

 the hydrolytic products of soap, i. e., alkali 

 or fatty acid, causes this lowering of the 

 surface tension; it is a measure of the 

 amount of soap present. Emulsification is 



