Section III, 188*7. [ 23 ] Teans. Koy. Soc. Canada. 



III.— Notes on the Analysis of Coffee. 

 By Anthony McGill, B.Sc, B.A. 



(Read and communicated by Dr. Macfarlane, May 27, 1887-) 



The followius criticisms of methods which obtain among food analysts are the result 

 of somewhat extensive work done upon commercial coti'ees at various periods during the 

 past two years. I have at times been led to suspect inherent faultiness in the recognised 

 methods, and in such cases have undertaken experimental research, which has occasion- 

 ally made very merciless demands upon my time. In the hope that the outcome of such 

 work may be found useful to other analysts, I have written this abstract, and requested 

 my friend, Mr. Macfarlane, to submit it to this Section of the Royal Society. 



The Extract or DecoctkiN. — This is made for the purpose of comparing one sample 

 of coffee with another sample, or with a standard which has been accepted by analysts. 

 Hence it is not so necessary that the whole of the soluble matter should be got into 

 solution as that the conditions under which the extract is made be such as to give (1) a 

 constant result, and (2) a result comparable with the accepted standard. 



The following methods are in use for the preparation of the extract : — 



{a.) Boiling the weighed substance with water for a definite time, throwing the 

 solution and residue upon a filter, and washing until the filtrate reaches a fixed volume. 



{b.) Boiling with the whole of the water to be used and, adapting a Liebig's con- 

 denser to the flask, continuing the boiling until a definite amount of distillate is obtained, 

 returning this to the flask, and filtering the solution. (Wanklyn.) 



(c.) Boiling with the whole of the water to be used and, attaching a back-flow con- 

 denser to the flask, continuing the boiling for a definite time. At the end of this time 

 the licjuid is filtered for use. (Wigner.) 



I have tried all these modes of making the extract, and find, that unless scrupulous 

 attention be paid to the minutest details of manipulation, duplicate extractions will give 

 decoctions differing so widely in density as to make that character practically unavailable 

 as a guide to tlie chicory percentage. With necessary attention to details, the last named 

 method gives excellent results. The following conditions must be observed : — 



I. — The coffee must be reduced to a standard degree of fineness. Thus, all other con- 

 ditions being alike, the following results were obtained (from Table III) : — 



Yorkshire chicory, coarse 1-02738 



Do. fine 1-02870 



Pure Mocha coffee, coarse 1-00849 



Do. fine 1-00978 



Pure .Java cofiee, very coarse 1-00862 



Do. coarse 100881 



Do. fine 1-00973 



