THE ANALYSIS OF GOLDEN SYRUP. BY NORMAN LEONARD, B.Sc. IN the December number of the Journal of the Society of Chemical I n d m t r y (1899, xviii., p. 1091) there appears, under the title of “Notes on the Analysis of Some Sngar Syrups,” a criticism, by Messrs. A. K. Miller, Ph.D., and J. H. Potts, of a recent paper on the analysis of golden syrup by Mr. R. Bodmer, Mr. H. M. Smith, and myself (ANALYST, 1899, xxiv., p. 253). Whilst admitting the fairness, from their own standpoint, of the objections raised to our work by Dr. Miller and his colleague, I wish to emphasize the fact that our paper was communicated to the Society of Public Analysts, and was written from the point of view of the Public Analyst; it was not in any way intended as an introduction of new processes for the exact analysis of sugar syrups.At the time we wrote, the analysis of golden syrup was new to Public Analysts, qud Public Analysts, and since numerous inquiries addressed86 THE ANALYST. to one of us indicated a lack of information on the subject, we thought it useful to outline a scheme of analysis, the details of which might be altered or extended to suit individual preferences, and to show how, by a balancing of probabilities, a fairly safe and generally intelligible opinion as to the nature of a sample might be arrived at. It is doubtful if any Public Analyst would care to certify the amount of glucose added to golden syrup more accurately than t o the nearest 5 per cent., and even then, seeing the variable composition of glucose and golden syrups, the adoption of such expressions as ‘‘ about 50 per cent.of glucose syrup ” or ‘ I at least 70 per cent. of glucose syrup” is advisable. I t will thus be seen that the meanings attached to the word “ approximate ” by Messrs. Miller and Potts on the one hand, and by ourselves on the other, are rather different, and that consequently many of the objections raised to our analyses lose somewhat of their force. Messrs. Miller and Potts seem not to appreciate the position of the Public Analyst, who is bound by the form of certificate prescribed by the Sale of Food and Drugs Acts to give a quantitative expression of the composition of an article which he believes to be adulterated ; he is not permitted to ‘( be content with a qualitative test.” This being so, recourse must be had to the method of averages, so much deprecated by Dr.Miller and Mr. Potts, which those whose business it is to examine products of variable composition are often obliged to employ. An exact determina- tion of all the constituents of a sugar syrup would doubtless be of great interest to the chemist and to the manufacturer, but such an analysis would, as a rule, be quite unintelligible both to the magistrate and to the vendor charged with adulteration. The analytical results would therefore require to be expressed in terms of such comparatively familiar substances as cane-sugar syrup and glucose syrup, and this would again necessitate the use of averages, for it would be impossible to say accurately how the water, ash, and organic matter were distributed between the two syrups in a mixture.Respecting the errors introduced into the estimation of the water by neglecting the influence of the ash on the specific gravity of the syrup solution, I may observe that the amount of water present is, to the Public Analyst, of very little importance. We were aware of this source of error, but it may be pointed out that large amounts of ash are, I believe, rarely found except in crude, genuine syrups which would be passed without comment by a Public Analyst, whilst in those samples which are largely adulterated with glucose syrup the ash is usually low, and the resulting error, which here becomes of niore importance, is also small, and comparable with that inherent in the divisor process when applied to such complex mixtures as those in question. Although not claiming so much accuracy for our analyses as Dr.Miller appears to have thought was the case, it may be of interest to make some remarks on the minus values obtained for invert sugar in No. 4 sample and some others. These values, which we had understood were not uncommonly obtained in the indirect analysis of complex and impure syrups, had not escaped our notice, and we had followed the matter up even farther than our critics seem to have done. For although it is quite true, taking i = 0 in the equation K = i + 0*53x, that the maxi?nzLnz percentage of glucose (on the basis of the 53 factor) in No. 4 sample is 62.6, it isTHE ANALYST. 87 possible to look at the matter from another point of view. Thus, in the equation , if there is no invert sugar i must equal 0 and x (glucose) becomes 75.8; and this is the minimzm percentage of glucose on the basis of the 134" factor.This sample is not improbably adulterated with a glucose syrup having a lower reducing power than the average assumed by us, and we have here an example of the discrepancies inseparable from the use of averages, which, however, as already stated, must necessarily be employed. I t should be remembered that the Public Analyst, whilst making use of average analyses as a basis for calculation and comparison, is, or should be, fully aware of the greater or less amount of reliance to be placed on them, and is accustomed to regard his results from various points of view, knowing that the most elaborate and accurate analysis may lead him astray unless due regard is paid to modes of interpretation and to other circumstances than analytical details. The statements of the proximate composition of the mixed syrups given at the close of our paper (ANALYST, xxiv., p. 257) are to be regarded in this light, and, although more accurate methods are greatly to be desired, Messrs. Miller and Potts appear unable to suggest a process better adapted to the wants of the Public Analyst than that described by Mr. Bodmer, Mr. Smith, and myself. ~ 66.5s + 134z - 23.li 100