The two products (705 and 706) were evaporated under exactly the same conditions and to what appeared to the operator to be the same consistency. After cooling, however, it was apparent that while the body of the two finished products was apparently equal, the consistency of No. 706 was superior to No. 705 in that the former was smooth and creamy, whereas the latter had a somewhat irregular, lumpy appearance. This difference was doubtless due to the greater content of soluble solids in No. 706. The color of the two samples was identical. A mixture of one part of No. 703 and three parts of No. 704 when evaporated in the laboratory to the same consistency was identical in every way with No. 706. From the composition as stated in Table 1 it is apparent that the flavor and food value of the clear juice, which is sometimes discarded (represented in No. 703), are practically identical with the unconcentrated pulp as it passes through the cyclone. In fact, the only difference between the two appears to be about one-half per cent of insoluble matter. When the product is allowed to separate, it seems probable that this insoluble material as it rises in the mass has a tendency to act like a filter and carry up with it a large proportion of the bacteria and moulds present. The scale on which the work was done did not permit of sufficiently accurate measurement of the finished pulp to warrant the calculation of the loss in quantity caused by discarding the juice. From the composition of the pulps and of the raw material, however, it is apparent that this loss is practically proportional to the percentage of juice discarded. It is apparent, therefore, that the evaporation of the material just as it passes through the finisher will yield a product of the same color, of better consistency, in considerably greater quantity, and at practically the same proportionate expense of concentration as the evaporation of the residue after discarding the juice in accordance with the custom mentioned above. COMPOSITION OF TOMATO PULP[1] Whole Tomato Pulp The results obtained by the examination of 33 samples of whole tomato pulp are given in Table 2. The concentration of the samples varies from unconcentrated pulp as it runs from the cyclone to pulps of very heavy consistency. This table contains the data from which Tables 4 and 5 were calculated, although during the season a partial analysis was made of a large number of other samples, and the data secured therefrom were in all respects confirmatory of the relations calculated from Table 2. [1] This chapter is taken largely from an article by Bigelow and Fitzgerald, published in the Journal of Industrial and Engineering Chemistry, 1915, vol. 7, page 602.