as known in old days. There
were many officers so adorned who made excellent gamekeepers or gillies,
and in particular I remember a certain stalwart major whose beard grew
in two inverted horns that splayed outwards on his chest, and who was
the very image of my father's old gardener. I once very nearly addressed
him as 'Horton' by mistake, for that happened to have been the
gardener's name.

CHAPTER VII

TO PHARI

The 'second advance' began in due course. The first few camping grounds
were small, so that we had to proceed on the three days' march to Phari
in several columns, two columns a day leaving Chumbi together, but
halting at separate camping grounds on the way up, and meeting again at
Phari.

This march to Phari was, until we actually reached the Phari plain,
quite the wettest I have known. It rained incessantly. The first day we
climbed a few miles up to Lingmatam. (How like one another the names of
places in this part of the world are! It took me months to distinguish
between Lingtam, Langram, and Lingmatam.) From Lingmatam (a sopping,
spongy, flat little plain nestling in the hills, that had obviously only
just missed its proper vocation of being a lake instead of a plain) we
marched up a rough bridle-path through pine-woods to Dhota. We had a
very long train of pack-mule transport in our column, and the checks up
that steep narrow winding path were interminable, while rain fell the
whole time. Whenever anything went wrong with a mule's load, which of
course happened frequently owing to the steepness and roughness of the
track, it was impossible to take the mule aside to adjust the load, for
there was no room at the side, and the mule had to be halted where he
was till the adjustment was completed. This involved the halting of say
five hundred mules, who happened to be behind the mule who had first
been halted. And when the latter at last moved off, it of course took an
appreciable interval of time before the next mule followed suit.
Multiply that appreciable interval by the number of mules in the rear,
say five hundred, and you find that it takes perhaps a full half-hour
before the five-hundredth is at last on the move again. Thus that
initial adjustment of a refractory load has cost the rear of the column
half an hour's delay, and by the end of the half-hour you may be sure
that the load of another mule has got loose, and the whole process has
to be repeated. This is just an instance of the tri