580 REPORT— 1901. 



in diameter Lad apparently no measurable influence in causing an earlier breaking 

 up of the stream-line flow. Although the electrothermal method of measuremen* 

 was quite satisfactory, it was found that the point of change was determined more 

 simply by placing the bulb of a sensitive mercury thermometer in the path of the 

 water as it emerged from the tube, and this had also the additional advantage of 

 showing the true temperature of the water. A glass prolongation, of slightly 

 greater diameter and connected carefully to the brass pipe by a specially con- 

 structed cone or adapter, enabled the reading of the thermometer to be observed. 

 It was a matter of considerable surprise to the authors to see the very sudden way in 

 which the reading of the thermometer indicated the point of change in the character 

 of the flow by an almost instantaneous change of reading. That the change in 

 the reading indicates the critical point was shown by introducing a colour band in 

 the ordinary way, in which case the band disappeared at the same moment the jump 

 in the thermometer thread took place. 



Since in the experiments the tube was heated on the outside, it might at first 

 sight appear that the temperature difference between the layers of water in direct 

 contact and the central column might produce a disturbing action on the flow, but 

 as this temperature difference was always small, the total jump in the thermometer 

 being seldom over a few tenths of a degree, the disturbance, if any, was reduced 

 to a minimum. Moreover, special experiments were repeatedly made to determine 

 a possible disturbing effect by maintaining the temperature of the walls of the 

 tube at different points above and below the water in the tube, but none could be 

 detected. 



It was necessary to have only a few degrees difference in temperature between 

 the walls of the tube and inflow water to obtain a measurable reading. 



Description of the A])2)aratus. 



We were fortunate in having at our disposal, through the kindness of Dean 

 Bovey, the facilities afibrded by the hydraulic laboratory, where the large experi- 

 mental tank, 20 feet high and 25 square feet in area, served admirably for a 

 reservoir. The tank stood on the bed rock, and was therefore free from vibration 

 or disturbance, and after the eddies had died out, occasioned by filling, the water 

 was in as completely quiet a state as possible. The water used for the experi- 

 ments was supplied from the Montreal mains, and was quite clear. It would not 

 have been possible to use distilled water owing to the large quantity required, but 

 every precaution was taken in the way of repeated cleaning to have the water 

 pure. 



The rest of the apparatus was designed, and for the most part constructed, in 

 the laboratory, and served admirably for fulfilling the required conditions for 

 carrying out the experiments. Subsequently it was found that by a few simple 

 alterations tlie method of colour bands could be used as well for the experiments 

 witli the large pipes. 



Each of the metal pipes studied was fitted with a metal trumpet flare to direct 

 the flow as it entered, the point of junction being very carefully smoothed so as to 

 produce no disturbing action. The walls of these pipes were maintained at a con- 

 stant temperature, above or below the temperature of the water flowing through, 

 by means of a jacket, through which water was circulated by a centrifugal pump. 

 A graduated valve regulated the flow, which was caught and measured in an 

 accurately calibrated copper measure. 



E.vperiniental Results. 



Two tables are given, the first showing the effect produced by increasing the 

 head of water in slightly increasing the critical velocity; and the second, the 

 effect of temperature between 15° and 86° C. These experiments were made with 

 the 0'414-iuch brass pipe. 



Two other tables are given, one showing the agreement of the observations of 

 Eeynolds by the method of colour bands with thoise of the Authors, when reduced 



