Eighteen Months in a Sealed Office Jar: A Teardown

The original build

The vessel was a four-litre clip-top preserving jar with an intact rubber gasket, twenty-two centimetres tall and eighteen across at the shoulder. Base layer: two centimetres of expanded clay, a fibre separator, then four centimetres of a coir, grit and bark mix.

Hardscape was three pieces of the same dark slate, set as a low ridge across the rear third with the largest piece leaning back into the slope. Substrate was raised at the back to about six centimetres and thinned to two at the front.

Planting was five species: Leucobryum glaucum in two cushions on the ridge, Hypnum cupressiforme carpeting the mid-ground, Thuidium delicatulum in the left corner, a small Fissidens patch at the lowest point, and three tiny Selaginella kraussiana plugs at the front. A pinch of springtails went in with the substrate.

Conditions over eighteen months

The jar sat on a filing cabinet four metres from a north-facing window, receiving no direct sun at any hour. Overhead fluorescent panels ran roughly nine hours on weekdays and not at all at weekends. Measured light at the glass averaged around four hundred lux during working hours.

Office temperature was heavily managed: nineteen to twenty-three degrees on weekdays, drifting to sixteen at weekends and holidays. This narrow, cool band turned out to be the build's greatest advantage.

The lid was never opened. No water was added at any point. The jar was wiped externally perhaps six times over the period, by whoever was passing.

What thrived

Hypnum was the clear winner. It began as a patch of perhaps forty square centimetres and by month eighteen covered most of the mid-ground and had climbed the lower slate, roughly tripling its area. Colour stayed a good mid-green throughout.

Fissidens, planted at the wettest point, expanded steadily and produced the deepest green in the jar. It also colonised the shaded underside of the leaning slate, which was the single most attractive thing in the finished build and was entirely unplanned.

Thuidium held its position without expanding. It looked slightly paler at eighteen months than at planting but showed no dieback. Given the light level, holding steady counts as success.

What failed and why

Both Leucobryum cushions were dead by month nine. They browned from the crown downwards starting around month four, and the pattern was unmistakable: permanent saturation with no drying cycle in a fully sealed jar. Leucobryum wants humidity, not standing wetness at the crown, and a gasketed lid gave it no relief.

The Selaginella etiolated badly. In four hundred lux it stretched, produced long pale internodes, flopped forward and rotted where it touched the substrate. Two plugs were gone by month six and the third persisted as a thin straggle.

Both failures share a cause: species chosen for how they look rather than for the conditions the vessel would actually deliver. Nothing about the outcome was surprising in retrospect, which is precisely the point.

The cleanup crew and the substrate

Springtails established and were visible on the glass at every inspection. When the Leucobryum died, the dead tissue was reduced to nothing over about six weeks without a mould bloom, which is a good demonstration of what a working crew is for.

The substrate had settled by roughly fifteen per cent, almost all of it in the first four months. The rear slope lost about a centimetre of height, which flattened the composition slightly but did not slump.

Opening the jar at eighteen months produced no sour smell — earthy, damp, unremarkable. The drainage reservoir held perhaps a centimetre of clear water, which is close to where it started.

What transfers to your build

In a fully sealed vessel, choose species that come from permanently saturated habitat. Pleurocarps and Fissidens are the reliable choices; cushion mosses need a partial seal that lets the crown dry back between wettings.

Four hundred lux for nine hours a day is enough for moss and not enough for vascular plants. If your light is that marginal, plant moss only and let the composition come from hardscape and topography instead of foliage variety.

Cool and stable beats warm and bright for temperate species. The office thermostat, not the light, is the reason this jar survived eighteen months without a single intervention.

Extended notes

• Weekend darkness had no observable effect. Mosses tolerate intermittent photoperiods far better than most houseplants.
• The colonisation of the slate underside took about a year to become visible and was the only genuinely unexpected event in the whole build.
• If this jar were rebuilt today, the changes would be: no Leucobryum, no Selaginella, and one more piece of slate to replace the visual weight the cushions were meant to provide.

Frequently asked