OF MILK: CORRECTION FACTORS AND THE INFLUENCE OF STIRRING: I1 h I11 225 It was originally intended to employ four different amounts of supercooling for each sample of milk, vix. about 1.5, 1.0,0.8, and 0.5" C. It was found, however, that the differences in the readings of the thermometer corresponding to these different amounts of supercooling were in the neighbourhood of 0.002"-an amount so small that the unavoidable errors of observation might possibly approach the same order. It was, therefore, decided to employ only two different amounts of supercooling, as widely apart as practicable, say about 1.5" and 04", and to increase the number of samples examined. It was stated by Hortvet that, unless a much greater amount of supercooling than 0.5" is employed in the use of his cryoscope and technique, the rise of the mercury column is not sufficiently pronounced, and that there is more or less wavering, so that difficulty arises in deciding on the exact point at which the top of the column becomes stationary. Elsdon and Stubbs (Eoc.it.)^ found the same result when using a supercooling of less than about 0.8"; the mercury rose very slowly, and did not maintain a steady position for any appreciable time, and the proper freezing-point might not, under these circumstances, be attained. In the first experiments, made with the object of ascertaining the super- cooling correction, it was observed that where the amount of supercooling was small, that is, less than say 0-75", even in the absence of alcohol in the jacket surrounding the freezing-tube, the rise of temperature, when freezing occurred, was very slow-so slow, indeed, as to suggest doubts whether the thermometer would indicate the freezing-point of the milk, influenced only by supercooling, owing to the reading being affected by the length of time which elapses and the possibility of imperfect thermal insulation, causing a nett loss of heat from the freezing tube and contents.To give an instance; in an experiment when the supercooling was 0-71" the time required for the mercury column to rise until, on observation through the telescope with the aid of the horizontal cross-wire, the ascent became imperceptible, was 92 minutes, as compared with about 3& minutes when the same milk was super- cooled 1-49'. It will be seen later that, in the absence of alcohol in the space surrounding the freezing-tube, the heat insulation of the milk is not perfect; a nett loss of heat occurs, for it is possible t o carry through a freezing-point determination under such conditions.The difference between the two methods of working- with and without alcohol-results in a longer time being required for the cooling of the milk when the space around the freezing-tube is occupied by air. There were also the difficulties of judging when the rising column of mercury had reached the highest point, due to the very slow movement which occurred in that region, and of getting satisfactory readings, owing to the short time that elapsed before a fall took place. The idea of following strictly the Hortvet technique, as regards stirring, for these particular experiments was therefore abandoned.In the first series of experiments, the results of which are recorded in Table 111, four or five stirrings of three strokes each were employed, and in the second series the mechanical stirrer, working at the rate of 40 strokes per minute, was in operation all the time, the thermometer rising until tapping became necessary. It does not appear that these variations in stirring caused any significant differences in the results obtained, but the times of rising of the mercury, after two different extents, large and small,OF MILK: CORRECTION FACTORS AND THE INFLUENCE OF STIRRING: I1 h I11 225 It was originally intended to employ four different amounts of supercooling for each sample of milk, vix.about 1.5, 1.0,0.8, and 0.5" C. It was found, however, that the differences in the readings of the thermometer corresponding to these different amounts of supercooling were in the neighbourhood of 0.002"-an amount so small that the unavoidable errors of observation might possibly approach the same order. It was, therefore, decided to employ only two different amounts of supercooling, as widely apart as practicable, say about 1.5" and 04", and to increase the number of samples examined. It was stated by Hortvet that, unless a much greater amount of supercooling than 0.5" is employed in the use of his cryoscope and technique, the rise of the mercury column is not sufficiently pronounced, and that there is more or less wavering, so that difficulty arises in deciding on the exact point at which the top of the column becomes stationary.Elsdon and Stubbs (Eoc. it.)^ found the same result when using a supercooling of less than about 0.8"; the mercury rose very slowly, and did not maintain a steady position for any appreciable time, and the proper freezing-point might not, under these circumstances, be attained. In the first experiments, made with the object of ascertaining the super- cooling correction, it was observed that where the amount of supercooling was small, that is, less than say 0-75", even in the absence of alcohol in the jacket surrounding the freezing-tube, the rise of temperature, when freezing occurred, was very slow-so slow, indeed, as to suggest doubts whether the thermometer would indicate the freezing-point of the milk, influenced only by supercooling, owing to the reading being affected by the length of time which elapses and the possibility of imperfect thermal insulation, causing a nett loss of heat from the freezing tube and contents.To give an instance; in an experiment when the supercooling was 0-71" the time required for the mercury column to rise until, on observation through the telescope with the aid of the horizontal cross-wire, the ascent became imperceptible, was 92 minutes, as compared with about 3& minutes when the same milk was super- cooled 1-49'. It will be seen later that, in the absence of alcohol in the space surrounding the freezing-tube, the heat insulation of the milk is not perfect; a nett loss of heat occurs, for it is possible t o carry through a freezing-point determination under such conditions.The difference between the two methods of working- with and without alcohol-results in a longer time being required for the cooling of the milk when the space around the freezing-tube is occupied by air. There were also the difficulties of judging when the rising column of mercury had reached the highest point, due to the very slow movement which occurred in that region, and of getting satisfactory readings, owing to the short time that elapsed before a fall took place. The idea of following strictly the Hortvet technique, as regards stirring, for these particular experiments was therefore abandoned. In the first series of experiments, the results of which are recorded in Table 111, four or five stirrings of three strokes each were employed, and in the second series the mechanical stirrer, working at the rate of 40 strokes per minute, was in operation all the time, the thermometer rising until tapping became necessary.It does not appear that these variations in stirring caused any significant differences in the results obtained, but the times of rising of the mercury, after two different extents, large and small,OF MILK: CORRECTION FACTORS AND THE INFLUENCE OF STIRRING: I1 h I11 225 It was originally intended to employ four different amounts of supercooling for each sample of milk, vix. about 1.5, 1.0,0.8, and 0.5" C. It was found, however, that the differences in the readings of the thermometer corresponding to these different amounts of supercooling were in the neighbourhood of 0.002"-an amount so small that the unavoidable errors of observation might possibly approach the same order.It was, therefore, decided to employ only two different amounts of supercooling, as widely apart as practicable, say about 1.5" and 04", and to increase the number of samples examined. It was stated by Hortvet that, unless a much greater amount of supercooling than 0.5" is employed in the use of his cryoscope and technique, the rise of the mercury column is not sufficiently pronounced, and that there is more or less wavering, so that difficulty arises in deciding on the exact point at which the top of the column becomes stationary. Elsdon and Stubbs (Eoc. it.)^ found the same result when using a supercooling of less than about 0.8"; the mercury rose very slowly, and did not maintain a steady position for any appreciable time, and the proper freezing-point might not, under these circumstances, be attained.In the first experiments, made with the object of ascertaining the super- cooling correction, it was observed that where the amount of supercooling was small, that is, less than say 0-75", even in the absence of alcohol in the jacket surrounding the freezing-tube, the rise of temperature, when freezing occurred, was very slow-so slow, indeed, as to suggest doubts whether the thermometer would indicate the freezing-point of the milk, influenced only by supercooling, owing to the reading being affected by the length of time which elapses and the possibility of imperfect thermal insulation, causing a nett loss of heat from the freezing tube and contents. To give an instance; in an experiment when the supercooling was 0-71" the time required for the mercury column to rise until, on observation through the telescope with the aid of the horizontal cross-wire, the ascent became imperceptible, was 92 minutes, as compared with about 3& minutes when the same milk was super- cooled 1-49'.It will be seen later that, in the absence of alcohol in the space surrounding the freezing-tube, the heat insulation of the milk is not perfect; a nett loss of heat occurs, for it is possible t o carry through a freezing-point determination under such conditions. The difference between the two methods of working- with and without alcohol-results in a longer time being required for the cooling of the milk when the space around the freezing-tube is occupied by air.There were also the difficulties of judging when the rising column of mercury had reached the highest point, due to the very slow movement which occurred in that region, and of getting satisfactory readings, owing to the short time that elapsed before a fall took place. The idea of following strictly the Hortvet technique, as regards stirring, for these particular experiments was therefore abandoned. In the first series of experiments, the results of which are recorded in Table 111, four or five stirrings of three strokes each were employed, and in the second series the mechanical stirrer, working at the rate of 40 strokes per minute, was in operation all the time, the thermometer rising until tapping became necessary.It does not appear that these variations in stirring caused any significant differences in the results obtained, but the times of rising of the mercury, after two different extents, large and small,OF MILK: CORRECTION FACTORS AND THE INFLUENCE OF STIRRING: I1 h I11 225 It was originally intended to employ four different amounts of supercooling for each sample of milk, vix. about 1.5, 1.0,0.8, and 0.5" C. It was found, however, that the differences in the readings of the thermometer corresponding to these different amounts of supercooling were in the neighbourhood of 0.002"-an amount so small that the unavoidable errors of observation might possibly approach the same order. It was, therefore, decided to employ only two different amounts of supercooling, as widely apart as practicable, say about 1.5" and 04", and to increase the number of samples examined.It was stated by Hortvet that, unless a much greater amount of supercooling than 0.5" is employed in the use of his cryoscope and technique, the rise of the mercury column is not sufficiently pronounced, and that there is more or less wavering, so that difficulty arises in deciding on the exact point at which the top of the column becomes stationary. Elsdon and Stubbs (Eoc. it.)^ found the same result when using a supercooling of less than about 0.8"; the mercury rose very slowly, and did not maintain a steady position for any appreciable time, and the proper freezing-point might not, under these circumstances, be attained. In the first experiments, made with the object of ascertaining the super- cooling correction, it was observed that where the amount of supercooling was small, that is, less than say 0-75", even in the absence of alcohol in the jacket surrounding the freezing-tube, the rise of temperature, when freezing occurred, was very slow-so slow, indeed, as to suggest doubts whether the thermometer would indicate the freezing-point of the milk, influenced only by supercooling, owing to the reading being affected by the length of time which elapses and the possibility of imperfect thermal insulation, causing a nett loss of heat from the freezing tube and contents.To give an instance; in an experiment when the supercooling was 0-71" the time required for the mercury column to rise until, on observation through the telescope with the aid of the horizontal cross-wire, the ascent became imperceptible, was 92 minutes, as compared with about 3& minutes when the same milk was super- cooled 1-49'.It will be seen later that, in the absence of alcohol in the space surrounding the freezing-tube, the heat insulation of the milk is not perfect; a nett loss of heat occurs, for it is possible t o carry through a freezing-point determination under such conditions. The difference between the two methods of working- with and without alcohol-results in a longer time being required for the cooling of the milk when the space around the freezing-tube is occupied by air. There were also the difficulties of judging when the rising column of mercury had reached the highest point, due to the very slow movement which occurred in that region, and of getting satisfactory readings, owing to the short time that elapsed before a fall took place.The idea of following strictly the Hortvet technique, as regards stirring, for these particular experiments was therefore abandoned. In the first series of experiments, the results of which are recorded in Table 111, four or five stirrings of three strokes each were employed, and in the second series the mechanical stirrer, working at the rate of 40 strokes per minute, was in operation all the time, the thermometer rising until tapping became necessary. It does not appear that these variations in stirring caused any significant differences in the results obtained, but the times of rising of the mercury, after two different extents, large and small,OF MILK: CORRECTION FACTORS AND THE INFLUENCE OF STIRRING: I1 h I11 225 It was originally intended to employ four different amounts of supercooling for each sample of milk, vix. about 1.5, 1.0,0.8, and 0.5" C.It was found, however, that the differences in the readings of the thermometer corresponding to these different amounts of supercooling were in the neighbourhood of 0.002"-an amount so small that the unavoidable errors of observation might possibly approach the same order. It was, therefore, decided to employ only two different amounts of supercooling, as widely apart as practicable, say about 1.5" and 04", and to increase the number of samples examined.It was stated by Hortvet that, unless a much greater amount of supercooling than 0.5" is employed in the use of his cryoscope and technique, the rise of the mercury column is not sufficiently pronounced, and that there is more or less wavering, so that difficulty arises in deciding on the exact point at which the top of the column becomes stationary. Elsdon and Stubbs (Eoc. it.)^ found the same result when using a supercooling of less than about 0.8"; the mercury rose very slowly, and did not maintain a steady position for any appreciable time, and the proper freezing-point might not, under these circumstances, be attained. In the first experiments, made with the object of ascertaining the super- cooling correction, it was observed that where the amount of supercooling was small, that is, less than say 0-75", even in the absence of alcohol in the jacket surrounding the freezing-tube, the rise of temperature, when freezing occurred, was very slow-so slow, indeed, as to suggest doubts whether the thermometer would indicate the freezing-point of the milk, influenced only by supercooling, owing to the reading being affected by the length of time which elapses and the possibility of imperfect thermal insulation, causing a nett loss of heat from the freezing tube and contents.To give an instance; in an experiment when the supercooling was 0-71" the time required for the mercury column to rise until, on observation through the telescope with the aid of the horizontal cross-wire, the ascent became imperceptible, was 92 minutes, as compared with about 3& minutes when the same milk was super- cooled 1-49'.It will be seen later that, in the absence of alcohol in the space surrounding the freezing-tube, the heat insulation of the milk is not perfect; a nett loss of heat occurs, for it is possible t o carry through a freezing-point determination under such conditions. The difference between the two methods of working- with and without alcohol-results in a longer time being required for the cooling of the milk when the space around the freezing-tube is occupied by air. There were also the difficulties of judging when the rising column of mercury had reached the highest point, due to the very slow movement which occurred in that region, and of getting satisfactory readings, owing to the short time that elapsed before a fall took place.The idea of following strictly the Hortvet technique, as regards stirring, for these particular experiments was therefore abandoned. In the first series of experiments, the results of which are recorded in Table 111, four or five stirrings of three strokes each were employed, and in the second series the mechanical stirrer, working at the rate of 40 strokes per minute, was in operation all the time, the thermometer rising until tapping became necessary. It does not appear that these variations in stirring caused any significant differences in the results obtained, but the times of rising of the mercury, after two different extents, large and small,OF MILK: CORRECTION FACTORS AND THE INFLUENCE OF STIRRING: I1 h I11 225 It was originally intended to employ four different amounts of supercooling for each sample of milk, vix.about 1.5, 1.0,0.8, and 0.5" C. It was found, however, that the differences in the readings of the thermometer corresponding to these different amounts of supercooling were in the neighbourhood of 0.002"-an amount so small that the unavoidable errors of observation might possibly approach the same order. It was, therefore, decided to employ only two different amounts of supercooling, as widely apart as practicable, say about 1.5" and 04", and to increase the number of samples examined. It was stated by Hortvet that, unless a much greater amount of supercooling than 0.5" is employed in the use of his cryoscope and technique, the rise of the mercury column is not sufficiently pronounced, and that there is more or less wavering, so that difficulty arises in deciding on the exact point at which the top of the column becomes stationary.Elsdon and Stubbs (Eoc. it.)^ found the same result when using a supercooling of less than about 0.8"; the mercury rose very slowly, and did not maintain a steady position for any appreciable time, and the proper freezing-point might not, under these circumstances, be attained. In the first experiments, made with the object of ascertaining the super- cooling correction, it was observed that where the amount of supercooling was small, that is, less than say 0-75", even in the absence of alcohol in the jacket surrounding the freezing-tube, the rise of temperature, when freezing occurred, was very slow-so slow, indeed, as to suggest doubts whether the thermometer would indicate the freezing-point of the milk, influenced only by supercooling, owing to the reading being affected by the length of time which elapses and the possibility of imperfect thermal insulation, causing a nett loss of heat from the freezing tube and contents.To give an instance; in an experiment when the supercooling was 0-71" the time required for the mercury column to rise until, on observation through the telescope with the aid of the horizontal cross-wire, the ascent became imperceptible, was 92 minutes, as compared with about 3& minutes when the same milk was super- cooled 1-49'. It will be seen later that, in the absence of alcohol in the space surrounding the freezing-tube, the heat insulation of the milk is not perfect; a nett loss of heat occurs, for it is possible t o carry through a freezing-point determination under such conditions. The difference between the two methods of working- with and without alcohol-results in a longer time being required for the cooling of the milk when the space around the freezing-tube is occupied by air. There were also the difficulties of judging when the rising column of mercury had reached the highest point, due to the very slow movement which occurred in that region, and of getting satisfactory readings, owing to the short time that elapsed before a fall took place. The idea of following strictly the Hortvet technique, as regards stirring, for these particular experiments was therefore abandoned. In the first series of experiments, the results of which are recorded in Table 111, four or five stirrings of three strokes each were employed, and in the second series the mechanical stirrer, working at the rate of 40 strokes per minute, was in operation all the time, the thermometer rising until tapping became necessary. It does not appear that these variations in stirring caused any significant differences in the results obtained, but the times of rising of the mercury, after two different extents, large and small,