No reliable conclusion, as to whether a particular specimen of milk has been adulterated or not, can therefore be drawn from the differences in specific gravity alone. A radical difficulty attending this test arises from the fact that the specific gravity both of water and cream is less than that of pure milk. If, therefore, the hydrometer sinks deeper into the fluid than would be expected in ordinary pure milk, how is it possible, unless the variation is very large, to tell whether it is due to the richness of the milk in cream, or to the water? I have, for instance, two instruments, each labelled “Lactometer,” but both of which are simple hydrometers (Fig. 71), or specific gravity testers, one of which is graduated with the water-mark 0 and that of pure milk 20°; the water-mark of the other being 0, like the first, and that of pure milk 100°. Both are the same in principle, the only difference being in the graduation. On the former, graduated for pure milk at 20°, it is difficult to tell with accuracy the small variations in the percentage of water or cream, the divisions on the scale are so minute, while the latter marks them so that they can be read off with greater ease and precision. For the purpose of showing the difference in the specific gravity in different specimens of pure milk, taken from the cows in the morning, and allowed to cool down to about 60°, I used the latter instrument with the following results: The first pint drawn from a native cow stood at 101°, the scale being graduated at 100° for pure milk. The last pint of the same milking, being the strippings of the same cow, stood at 86°. The mixture of the two pints stood at about 93¹⁄₂°. The milk of a pure-bred Jersey stood at 95°, that of an Ayrshire at 100°, that of a Hereford at 106°, that of a Devon at 111°, while a thin cream stood at 66°. All these specimens of milk were pure, and milked at the same time in the morning, carefully labelled in separate vessels, and set upon the same shelf to cool off; and yet the variations of specific gravity amounted to 25°, or, taking the average quality of the native cows’ milk at 93¹⁄₂°, the variations amounted to 17¹⁄₂°. But, knowing the specific gravity, at the outset, of any specimen of milk, the hydrometer would show the amount of water added. This cheap and simple instrument is therefore of frequent service. The lactometer is a very different instrument, and measures the comparative richness of different specimens of milk. It is of very great service both in the butter and cheese dairy, for testing the comparative value of different cows for the purposes for which they are kept. This instrument is very simple and cheap and the practical dairyman can tell by it what cows he can best part with without detriment to his business. No cow should be admitted to a herd kept for butter-making without knowing her qualities in this respect.