A man with a plan: Take the path to enlightenment in the New Year

My great uncle Douglas was a meticulous man: tall, gaunt, gaitered and solitary. His intellect was terrifying. The sitting room of the cottage he lived in was lined entirely with books. He had, I realised when I had to dismantle the place after his death, built up an enormous library of 20th-century thought.

Great uncle Douglas's life as a schoolmaster left him plenty of time to garden, which he did with the same extraordinary attention to detail that characterised all his actions. He kept notes, of course, in ruled books with stiff marble covers, recording the progress on his acre plot. Opening one at random, I find "Plan of Garden for 1954" written at the top of a page in his handsome schoolmasterly script. Underneath is his scheme for the vegetable plot. On the left hand side of the path, 'Pilot' peas, to be followed by two rows of Brussels sprouts and one row of broccoli. Below the peas, blackcurrants, then three rows of Savoy cabbage. After that 'Onward' peas, lettuce, radish and seedbeds for the sprouts and broccoli. Finally turnips, beetroot, parsnips and onions with an edging at the far end of strawberries. In July, I was allowed to scavenge there.

His plans for the right-hand side of the path are equally prosaic: masses of potatoes, broad beans, kale, a very Fifties diet. But he grew flowers too, not jumbled into a herbaceous border but divided into blocks of his specialities carnations, auriculas, sweet peas. Fifty years ago, in the New Year of 1958, he records planting a new orchard with 'Devonshire Quarrenden', 'Brownlees Russet' and 'Adams's Pearmain' apples. All are still available and, if I were planting an orchard this year, I'd take his selection straight from the page. He would have spent more time than most gardeners sizing up the different varieties, working out pollination tables and planning for the longest possible succession of good, keeping fruit.

His fastidious care makes me all the more aware at the beginning of this new year that my own Plan of Garden does not exist. All I have in my head is a hazy list of things that I'd like to see done over the next season. You note "see done" rather than "do". Unfortunately, most of the things on my mind are things I can't actually accomplish myself. Like the path that we need, to join the bottommost grass terrace with the one above it. It's a steep bank, the upper terrace four and a half metres above the lower one. Fortunately, it's not ever going to be a major route. It's a snicket, like the alleys that connect the main parallel streets in old towns like Stamford, never meant to bear more than feet, certainly not a wheelbarrow.

So it can be narrow. But it's got to be stable, because of the steepness. We can't just carve out a route and then top dress it with chipped wood or gravel, as we have with the wider paths in the main garden. Effectively, there's only one place where we can connect the two levels and allow the path to take a line across the slope rather than straight up it. But it has to get by a big old willow, a 'Kiftsgate' rose and a monumental Viburnum plicatum that is one of the best things we inherited in the garden.

When we arrived, the 'Kiftsgate' rose was flinging its great arms into the willow and in bloom, looked fantastic. But willows are brittle things and in one of the winter storms that came surging up the valley, some of the willow branches crashed down, bringing the rose with them. Fortunately, it was the right time of year to be able to chop 'Kiftsgate' down to the base and start it off again, training it this time on three totem poles. In a previous life, they were newel posts from a stair we took out of the house. They're oak, so I hope they'll last.

A remnant of a path is there already, made, I expect, for the same reason: a need to connect two levels of the garden and an understanding that there were few possible options. This earlier path is made of biggish pieces of stone. But not big enough to stay in place. By the time we arrived, most of them had slipped down the hill and the path itself was well disguised under the surrounding greenery.

I think stone is indeed the best option, set so that one side of the stone becomes the riser of the step and the adjoining face the tread. This calls for fairly hefty, roughly rectangular blocks not the kind of stone that is naturally found round here. Rough field rubble is what cottages round here are built from. The stones of the original path must have slipped because they weren't big enough.

I'm wondering whether some hamstone blocks we've got might do the job. Originally they faced the garage that the previous owners of the house built. We took it down, for the sake of the view beyond, but saved the stone. The blocks have been properly quarried and some of them are at least 40cm by 25cm. Well bedded-in, side by side, will they stay where they are put? The only person who can decide is Kevin, who works here every Thursday. Fortunately, compared with the mammoth task of making the rammed stone path through the flower garden at five feet wide and 84 feet long, this one seems a mere doodle.

So, the willow path is, I think, the priority, more important than the floor of the hammock house, which is still bare, beaten earth. One day I'll have the time to draw out a pattern on the earth and gather together the bits and pieces I'll need to make the floor. We're running out of our own salvaged flotsam. That's a strange feeling. All my gardening life we've had hoards of brick, edging tile, slate, cobbles and timber to draw on. A friend reminded me of the beautiful floor in a summerhouse at Heligan, made from animal vertebrae. Nice, yes, but I'm not sure I fancy hanging round the abattoir to collect the goods themselves.