326 II. C. Bradley — Color Reaction for Copper. 



Art. XXX. — A Delicate Color Reaction for Copper, and a 

 Microchemical Test for Zinc / by Harold C. Bradley. 



[From the Chemical Laboratory of the Yale Medical School.] 



It has been known for a number of years that the extract 

 of logwood-haematoxylin — would produce with copper salts a 

 dark blue color of some intensity. One of the older and little 

 used differential stains for tissues depended upon this reac- 

 tion. The microtome section of tissue was immersed first in a 

 copper solution, then washed and immersed again in a dilute 

 solution of haematoxylin. Those portions of the tissue which 

 fixed the copper would then be colored dark blue, while the 

 rest of the tissue remained uncolored. In this way a stain is 

 produced which will differentiate the cell nucleus from the 

 surrounding cytoplasm. This same reaction was used by Herd- 

 man and Boyce* to demonstrate copper in the blood and tis- 

 sues of the oyster, while Mendel and Bradley f made use of it 

 in localizing the copper in the liver tissues of other marine 

 molluscs. So far as we are aware, however, the reaction has 

 never been used as a means of identifying small amounts of 

 copper in solution, nor has it been realized of what extreme 

 delicacy the reaction is susceptible. 



Accordingly a number of trials w T ere made with copper sul- 

 phate solutions of varying strengths, to determine within what 

 limits the reaction was available for the detection of copper, 

 and also how the reaction compared in delicacy with other 

 well known tests for that element. Ferrocyanide, ammonium 

 sulphide, potassium iodide and starch, are the reagents most 

 commonly employed to detect small amounts of copper, and 

 form some of the most delicate reactions of the laboratory. 

 Potassium ferrocyanide gives, with dilute solutions of copper 

 salts, a characteristic brown color, becoming indistinguishable 

 from the color of the reagent when the copper solution con- 

 tains less than O001 per cent of the metal. With starch 

 paste and potassium iodide the reaction is slightly more deli- 

 cate — cuprous iodide and starch iodide of characteristic deep 

 blue being formed — but reaches its limit when the copper 

 solution contains less than 0*001 per cent of the metal. On 

 the other hand, the haematoxylin reaction is at its best in just 

 such dilutions and will serve to recognize copper in solutions 

 of much greater attenuation. The following table shows roughly 

 the comparative delicacy of these reactions : 



Reagent 0-01$ Cu 0-001$ 0.0001$ 0-00001$ 0-000001$ 0'000000l$ 



Ferrocyanide brown brown ? 



KI + starch blue blue blue 



Haematoxylin blue blue blue blue blue blue 



* Herdman and Boyce : Report of the Thompson-Bates Laboratories, 

 Liverpool, ii, 1899. 



f Mendel and Bradley : American Journal of Physiology, xiv, 313, 1905. 



