430 H.A. Ø.Bistrup. 



the upper-structure on the quarter-deck, and from there over a 

 second pulley down alongside a measuring board divided into meters 

 and centimeters. Here at the free end of the wire was a heavy 

 lead-weight with a pointer, which could move up and down the 

 scale. 



The readings of the scale gave directly the height of the surface 

 above the fixed point. Now and then the reading was controlled by 

 a simultaneous reading of the height of water on a pole passed down 

 through the fire-hole alongside the wire. When the wire broke, which 

 happened fairly frequently, such control was necessary and then 

 by shortening or lengthening the wire agreement between the two 

 readings was reestablished. If the difference was very small, no 

 change was made in the wire, but a correction was then determined 

 which was at once applied to the reading. 



Thus, when the apparatus was in good working order, the verti- 

 cal movements of the ship could be constantly followed , hence the 

 course of the tidal wave, from centimeter to centimeter, and even 

 at high and low water, when the movements were small, the slightest 

 change reflected itself in the position of the pointer. 



The daily observations from October 3rd 1906 until the series 

 concluded on March 2nd 1907 are recorded in the accompanying 

 tables. Though the readings are given in centimeters, that does not 

 mean , that the fluctuations in the height of the water could be 

 determined with an accuracy of 1 cm., for the ship itself, however 

 fixed it seemed to be in the ice, has certainly had some amount of 

 independent movement in its bed. Similarly, we may assume, that 

 the ship has also had a slight amount of freedom in horizontal direc- 

 tion , Avhich , from the mode of construction of the apparatus , has 

 also had some influence on the up and down movements of the 

 pointer. 



During the observations special attention had to be paid to the 

 following conditions. 



When the vertical movement of the ship was very slight at low 

 and high water, it very frequently happened, that the wire became 

 frozen in the ice in the fire-opening and thus the pointer remained 

 standing at the same spot. It was thought in the beginning, that 

 this indicated a maximum or minimum point in the height of the 

 tidal wave, but the source of irregularity was soon discovered and 

 it was found necessary to test now and then whether the wire ran 

 freely through the fire-opening. When the wire became fast in the 

 ice, the reason was as a rule, that the fire-opening had not been 

 cut through that day, so that the funnel-shaped opening — as shown 

 in the sketch — had closed. But there was also a second reason. 

 For some time we used a wire composed of several strands. If one 

 or several of these sprung, they acted like drags and were more apt 



