EEPOET OF THE CHEMIST. lo7 



and remove. If tlie liquid is vigorously stirred a few moments, a greenish precipitate 

 Avill gather and a clear liquor remain ; if, on the other hand, not enough of the acid solu- 

 tion has been added, the liquor will be cloudy in appearance and contain but a sinal! 

 quantity of precipitate. If such is the case, dip a small splint in the acid solution and 

 add to the contents of tube till a good precipitate takes place. If too much acid solu- 

 tion has been used in the defecator, no precipitate will form in the tube, nor will the 

 albumen coagulate, a condition which is very bad, requiring the use of lime at an im- 

 proper time to correct. This tube test will give tlie exact state of the contents of the 

 delecat .c and by it the correct acid point obtained. 



*'Tlie heat is now raised till the contents of the defecator stand at a temperature oC 

 190° F., wlion it is shut off; wo now add milk of lime (which will form more precipi- 

 tate) till the liquor is exactly neutral; good, sharp, caustic lime should be used, as it will 

 take less of it, and our experience sliows that the less we have to use the better. 

 Another point at this stage is to .see that the liquors do not grow alkaline after they 

 show neutral, as lime acts somewhat slowly ; if such should be the case, add enough 

 of the acid solution till litmus changes to a purplish blue, but the neutral i^oint must 

 bo reached to cause a full precipitation so as not to be bothered with it during con- 

 centration. 



"After defecation we pass the juice into a vessel called a separator, of the same ca- 

 pacity as the defecator, and collect the precipitate in the bottom of the cone, from 

 which it is drawn ofi' through the faucet, thereby saving a great deal of labor in 

 straining; if the liquor should still show any cloud, then strain it through a filt»'r 

 press or bag filter, when it will be clear and brilliant, of a light lemon color." 



Unfortunately we did not have time to make a trial of tlie process 

 until after the cane had been injured by frost. I am of the opinion that 

 the process carried out as deacribed above will not Trork any injury to 

 the sucrose in the juice. 



The result of the defecatiou showed that the juice filtered through 

 filter paper was limpid and of a light lemon color. About 2,000 pounds 

 of it was sent to the filter press, but in spite of every endeavor it would 

 not pass through. In a few moments the cloths were compjetely closed 

 and at the highest ])ressuro obtainable no liquid could pass. The diffi- 

 culties with the Wilcox method are therefore not chemical but mechani- 

 cal. I cannot suggest any method by which these mechanical difficul- 

 ties can be overcome. 



In concluding this report I desire to properly thank Prof. M. A. 

 Scoveli and JMr. James Forsythe for the services they rendered in erect- 

 ing the machinery, and my assistants, Messrs. Cliilbrd Kichardson, G. 

 L. Spencer, and John Dugan, for the aid they gave in the laboratory 

 and in the factory. 



I desire also to thank Professor Sweuson and Mr. W. L. Parkinson, 

 the chemist and the manager of the Franklin Sugar Company, for the 

 many courtesies shown us by them, and for the readiness with which 

 they accorded us every facility of manipulation in their power. 



GENERAl, CONCLUSIONS. 



The general results of the experiments may be summarized as fol- 

 lows: 



(1) By the process of diffusion 98 per cent, of the sugar in the cane 

 was extracted, and the yield was fully double that obtained in the ordi- 

 nary way. 



(2) The difficulties to be overcome in the application of diffusion are 

 wholly mechanical. With the apparatus on hand, the following changes 

 are necessary in order to be able to work 120 tons of cane per day : 



(a) The diffusion cells should be made twice as large as they now are, 

 i.e., of 130 cubic feet capacity. 



(b) The opening through which the chips are discharged should be 

 made as nearly as possible of the same area as a horizontal cross-sec- 

 tio?. of the cell. 



