728 PROCEEDINGS OF THE AMERICAN ACADEMY. 



pass in opposite directions through the connectors, in which they can 

 be clamped by the screws. The end hooks are thus capable of being 

 raised or lowered or of being moved laterally, so that the combination 

 of the three hooks will support any tube and capillary. 



We found it best to protect the capillary from direct contact with 

 the flame, and at the same time to secure a more uniform heating by 

 enclosing it in a brass tube or collar which is slipped over the capillary 

 and rests on the two anterior hooks of the support. This collar is 

 5 cm. long, with an outer diameter of 6 mm. and an inner of 4 mm. 

 It is heated towards the anterior end by the tip of a flame 5 to 6 mm. 

 high from a burner with a good air supply and protected by a conical 

 chimney. We assume that under these conditions the capillary is 

 heated, through a space of 3 cm., at about 500°. 



The Preparation of Standard 3Iirrors. 



A standard solution of antimony was made by dissolving 2.3068 gr. 

 of recrystallized tartar emetic in water and making up to one liter. 

 This solution (I) contains 1 mg. of antimonious oxide in each cubic 

 centimeter. Of solution I, 10 c.c. were diluted to a liter, giving a 

 solution (II) containing 0.01 mg. per c.c. The tartrate offers the 

 most convenient and accurate solution, and we satisfied ourselves 

 that the presence of the tartrate ions had no effect whatever on the 

 deposition of the mirror. The strength of solution I was also checked 

 by analysis. 



The zinc used in the reduction flask was the same as that used in 

 the generator, but in smaller pieces, averaging perhaps 1 cm. in their 

 longest dimension. The weight used was from three to five grams. 

 This zinc, which is obtained of the New Jersey Zinc Company of New 

 York, contains not over 0.019 per cent of lead and not more than 

 0.013 per cent of iron ; hence the evolution of hydrogen by its contact 

 with dilute sulphuric acid is slow. Platinum or other sensitizing agents, 

 either in form of foil or as a deposit on the zinc, cannot be used, as we 

 have proved by trial, since they show a tendency to hold back anti- 

 mony even greater than in the case of arsenic. ^-^ We had recourse, 

 therefore, to hydrochloric acid, in a dilution of one to ten, of which we 

 use exactly 20 c.c. for each run. The hydrochloric acid, which is ob- 

 tained of Messrs. Baker and Adamson, of Easton, Pa., contains no 

 antimony, and only about 02 mg. of arsenious oxide per liter, an 



^^ The discussion as to tlie effect of otlier Jiietais on the evolution of arsenic 

 hych'ide has heen revived in recent years through the endeavors to increase tiie 

 delicacy of the Marsh test for arsenic. This question, as ai)i)lied to arsenic and 

 antimony, will he taken up hy one of us in a fulure jiaper. 



