March 15, 1906] 



NA TURE 



467 



fraud. If butter had always the same composition the 

 matter would be simple enough. It is the natural 

 variation, small though it be, which has hitherto 

 enabled the sophisticator to pocket his ill-gotten 

 fortune. 



This variation arises from a number of causes, of 

 which some at least have been fairly well elucidated. 

 In the main it is due to differences in the feeding and 

 treatment of the cows. For instance, Dr. J. J. L. 

 van Rijn showed some years ago in Holland that 

 with the approach of winter, when grass was be- 

 coming poorer and less plentiful-, and cows, kept late 

 in pasture, were much exposed to cold, the proportion 

 •of volatile acids in the butter became abnormally 

 low ; but that as soon as the animals were stabled, and 

 therefore better fed and protected from inclement 

 weather, the volatile acids began to increase. To 

 treat the cows more generously in the matters of 

 warmth and provender was thus the remedy for the 

 abnormality indicated. Nevertheless, there still re- 

 main some minor causes of variation, such, for 

 instance, as the different physiological conditions of 

 the cows in varying stages of lactation. The effects 

 of these, however, are largely neutralised when we 

 ■deal with butter made from the mixed milk of many 

 •cows ; though they have to be taken into account 

 where nothing is known about the history of a 

 particular specimen of butter. 



What, then, it may be asked, are the means by 

 which any check at all can be placed upon the 

 sophistication? Having regard to the variation in 

 admittedly genuine samples, is not analysis useless 

 except in very gross cases of fraud? By no means. 

 Let us see what it can do, and how it may be 

 supplemented. 



I n his search for adulterants the analyst may 

 determine the following constants of the butter- 

 fat he is examining : — (i) Its specific gravity : in 

 butter this is higher than in beef-fat, but lower than 

 in cocoa-nut oil; (2) its refractive index: this again 

 is intermediate in value between those of the two 

 adulterants mentioned; (3) its " saponification " value, 

 which gives a measure of the total quantity of fatty 

 acids and also of the glycerin ; and (4) most important 

 of all, the proportion of volatile acids. So far as 

 concerns butter, these physical and chemical data are 

 all correlated with the fact that glycerides of the 

 volatile acids form a relatively large proportion of 

 the genuine fat. They are therefore correlated with one 

 another, and, within limits, are interdependent. The 

 more volatile acid there is, the greater is the specific 

 gravity and the higher the saponification value, 

 whilst the refraction alters inversely. There is thus 

 a kind of parallelism preserved among the constants 

 of genuine butter, notwithstanding the variations in 

 their absolute values. For instance, given genuine 

 butter-fat such as that already referred to as " No. 2," 

 with 6-7 per cent, of volatile acids, one can predict 

 with confidence that its specific gravity will be pretty 

 close to 0.9130 (37°-8 C), and its saponification value 

 not much different from 232. On the other hand, if 

 the butter-fat contains, like No. 1, only 5 per cent, 

 of volatile acids, we can say with equal confidence 

 that its specific gravity will be about 0-9105, and its 

 saponification value approximately 222. 



Suppose, now. an adulterator mixes cocoa-nut oil with 

 the first of these butters until he has reduced the vola- 

 tile acids from 6-7 per cent, to 5 per cent., the quantity 

 in the second butter. Such an admixture would easily 

 be detected, notwithstanding the fact that the volatile 

 acids correspond in quantity with those of genuine 

 butter. For one thing, these acids can be further 

 examined and made to yield evidence of the admix- 

 ture; and for another, the addition of the cocoa-nut 



no. 189S, vor, 7 3 "I 



oil has destroyed the parallelism referred to ; the 

 figures are now quite inconsistent with one another, 

 and the sophistication is readily demonstrable. Speak- 

 ing generally, with a sufficiently extended analysis 

 there is no particular difficulty in detecting relatively 

 small quantities of cocoa-nut oil in butter. But a 

 mere determination of the proportion of volatile acids 

 will not suffice — which explains, perhaps, why this 

 special form of fraud has enjoyed a rather longer life 

 than it might otherwise have had. 



The addition of beef-fat or lard is somewhat more 

 difficult to prove. Reliance has chiefly, though not 

 entirely, been placed upon the consequent diminution 

 of the volatile acids. Unfortunately, the natural 

 variation of these in genuine butter is rather con- 

 siderable, as has been pointed out. Yet even so, it 

 does not follow that the lowest known limit for 

 volatile acids must be taken as the criterion in form- 

 ing a judgment. There are often collateral circum- 

 stances which narrow the range of admissible vari- 

 ation. For instance, a large consignment of 

 " creamery " butter must necessarily, under ordinary 

 commercial conditions, be a product of the mixed milk 

 of many cows, and hence the proportion of volatile 

 acids must tend towards the average value. The 

 minimum quantity, which in exceptional circumstances 

 milk-fat from a single cow might admittedly show, is 

 here lost in the general mean. No weight need be 

 given to it in judging the genuineness of the butter. 

 Similarly, samples purporting to be butter produced 

 in the summer or winter months cannot claim to be 

 judged by the low minimum sometimes found in 

 autumn butter. Again, it may be known, from 

 systematic analyses of genuine butter produced in a 

 certain region, and sampled in circumstances which 

 guarantee the representative character of the samples, 

 that the butter of this region in a specified month did 

 not, as a fact, fall below a particular value in the 

 matter of volatile acids. By having regard to these 

 and similar considerations, as well as to the actual 

 analytical figures, it has been possible in many cases 

 where butter adulterated with beef-fat or lard had been 

 imported into this country from the Continent, not 

 only to prove the fact of the adulteration, but to form 

 a fairly close estimate of its amount. 



Some check has thus been placed upon the fraud in 

 question. Yet, although the means at disposal will 

 serve to discover the more considerable amounts of 

 adulteration, there is still a residuum of cases which 

 are either doubtful or in which the legal proof is 

 difficult by reason of the clever way in which advan- 

 tage has been taken of the natural variation in the 

 constants of butter. To deal with these cases other 

 means are required in supplement of the chemist's 

 work. In Denmark, for instance, there is official 

 supervision of the butter industry ; and in Holland 

 the Government has organised an admirable system 

 of State " control," whereby the official imprimatur is 

 accorded to butter produced in factories under the 

 Government inspection. Joining the "control," how- 

 ever, is at present a voluntary matter, and for fac- 

 tories not under it, and which may be suspected of 

 malpractices, a compulsory and more rigid system 

 of inspection has been proposed. In this country a 

 considerable amount of falsification has been going 

 on. Enterprising and unscrupulous individuals import 

 or manufacture the adulterants mentioned above, and 

 offer, in consideration of a substantial fee, to initiate 

 the proprietors of butter-blending factories into the 

 whole art and mystery of butter-" faking." As 

 regards the suppression of this, the Customs authori- 

 ties in the first instance endeavour to secure that all 

 importations of adulterants coming within the legal 

 definition of margarine shall be marked accordingly. 



