POLAE YEAR EXPEDITION — STAGG 113 



look clocks which had been specially designed to function at low 

 temperatures ; we also had some lubricating oil alleged to have a very 

 low freezing point. But the oil froze in its bottle when put outside 

 the door in midwinter, and the low temperature clocks gave much 

 more trouble than, and were indeed ultimately discarded in favor of. 

 ordinary clocks used at meteorological office stations, after they had 

 been completely overhauled and cleaned of every trace of lubricant. 

 For future work where clocks are required to run continuously and 

 regularly at low temperatures, much more attention should be paid 

 to the design of the balance wheel and escapement movement. 



Our Dines anemometer erected over the main observing hut roof 

 Avorked very satisfactorily. Instead of galvanized iron tubing to 

 convey the pressure and suction surges from the head to the recorder, 

 we used hose piping. This stood up to the cold admirably and much 

 eased the jDrocess of erection. Only once or twice in the early days 

 of winter had we difficulty with the water of the recorder freezing. 

 A cupboard into which w-e could put a Valor heating stove was built 

 completely around it, and when the temperature of the hut showed 

 signs of falling below 32° F., the lighting of the stove insured safety, 

 A greater cause of trouble in the running of the recorder was the 

 drying of the pen. Throughout the winter months, when there was 

 probably no unfrozen water surface nearer than the Arctic Ocean, to 

 the north, absolute humidity generally was very low, and this, added 

 to the dry heat produced by a large iron box-stove burning wood 

 fuel, made the air in the hut so extremely dry that many ordinary 

 operations became difficult. In particular, the capillary action of the 

 surface of the anemograph record in drawing the ink from the pen 

 seemed almost to be prohibited. A great many remedies were tried, 

 but none was wholly satisfactory. 



The recording of snowfall has ils own pitfalls. Even to esti- 

 mate the total fall over a 24-hour period is difficult Avhere the snow 

 is fine, dry, and powdery, and so liable to drift with the slightest 

 wind. We had a continuously recording snow recorder of the Hell- 

 man-Fuess type as well as a rain gage converted into a snow gage, 

 and two or three snow poles distributed around our station, but the 

 days on which we got agreement between any two of these were 

 exceptionally few. An attempt was made to estimate the drift snow 

 by arranging a box with vertical circular opening and a system of 

 baffle plates to be automatically turned into the wind, but the con- 

 tributions to knowledge from this bit of our gear are probably 

 negligible. 



An important aspect of the meteorological program was our aero- 

 logical work, and for that we required a considerable stock of hydro- 

 gen. The very long and expensive overland route to our station 

 in Canada made the transport of hydrogen in cast steel cylinders 



