out of doors; he allows me to rub him and push him about without the slightest protest and scampers about me as I walk abroad. He is a strange beast—I imagine so unused to kindness that it took him time to appreciate it.
Tuesday, May 16.—The north wind continued all night but dropped this forenoon. Conveniently it became calm at noon and we had a capital game of football. The light is good enough, but not much more than good enough, for this game.
Had some instruction from Wright this morning on the electrical instruments.
Later went into our carbide expenditure with Day: am glad to find it sufficient for two years, but am not making this generally known as there are few things in which economy is less studied than light if regulations allow of waste.
Electrical Instruments
For measuring the ordinary potential gradient we have two self-recording quadrant electrometers. The principle of this instrument is the same as that of the old Kelvin instrument; the clockwork attached to it unrolls a strip of paper wound on a roller; at intervals the needle of the instrument is depressed by an electromagnet and makes a dot on the moving paper. The relative position of these dots forms the record. One of our instruments is adjusted to give only 1⁄10th the refinement of measurement of the other by means of reduction in the length of the quartz fibre. The object of this is to continue the record in snowstorms, &c, when the potential difference
