One species, two varieties, and what that rank means
A botanical variety is a rank below species: a recognisably distinct population within a species, distinct enough to name and not distinct enough to separate. Camellia sinensis has two universally accepted ones. Var. sinensis is the small-leaf form of south-west China and its neighbours, a multi-stemmed shrub with narrow, hard, serrated leaves and good cold tolerance. Var. assamica is the large-leaf form of Assam and upper South-East Asia, a tree with broad glossy drooping leaves, vigorous, heat-tolerant and frost-tender. They interbreed freely, which is why they are varieties rather than species and why so much planted material is a mixture of them. This is the level above cultivar, and it is the level that most reliably predicts what a tea will be like. Yabukita and Longjing 43 are different cultivars and both are var. sinensis, and the family resemblance between what they produce is much stronger than the resemblance between either and a large-leaf Yunnan plant.
What actually differs, beyond leaf size
Leaf size is the visible difference and not the important one. The large-leaf form grows faster and yields more, tolerates heat and humidity, and dies in a hard frost; the small-leaf form grows slowly, yields less, and survives winters that would kill its relative. That divergence sets the map: the northern limit of tea cultivation is effectively a small-leaf limit, and the hot lowland tropics are effectively large-leaf territory. Compositionally the two differ in direction as well as degree. Large-leaf material carries more of the compounds associated with body, colour and briskness, which is why it suits full oxidation and why it stands up to milk. Small-leaf material is generally more aromatic and less tannic, which is why it dominates the categories where aroma carries the tea — green, white, oolong. Those are tendencies rather than laws, and processing can move a tea a long way in either direction, but the tendency is strong enough that the variety is the first thing worth knowing about an unfamiliar plant.
The third name
Specialists disagree. TeaHQ does not pick a winner.
There is a third form, from the southern Indochinese uplands, which is intermediate in leaf size and upright in habit. Its name has never settled. It has been published as Camellia sinensis var. cambodiensis, treated as a subspecies of a separate assamica species under the epithet lasiocalyx, and dismissed as not distinct enough to name. The Indian planting tradition sidestepped the argument by calling it the Cambod jat and using it, and a great deal of Indian and Sri Lankan planting material carries its ancestry. TeaHQ records the name at `conflicting` confidence because that is what the literature is. The wider point is that infraspecific ranks in a domesticated, freely interbreeding, long-cultivated plant are a genuinely hard problem rather than a solved one that some authors get wrong. Boundaries drawn between populations that have been crossed by people for centuries are always going to be partly conventional, and a reader who finds tea’s botanical vocabulary unstable is noticing something real about the plant rather than a failure of scholarship.
How the two varieties came to be recognised as one plant
TeaHQ's own reading of well-attested general knowledge, with no single source behind it.
The recognition that the large-leaf plants growing in Assam were tea, rather than a separate genus or a wild curiosity, took place during the early nineteenth century and was contested at the time. The stakes were commercial rather than academic: if tea could be grown in British-held territory it need not be bought from China, and the botanical question had an empire’s trade balance resting on it. Chinese seed and Chinese plants were imported into India alongside the indigenous material, which is how the two varieties came to be growing side by side and hybridising. TeaHQ presents this as its own reading of general reference material rather than as archival history, and the specific dates, individuals and priority claims that circulate around the episode are not asserted here — they are among the most heavily repeated and least primary-sourced passages in tea writing. What is not in question is the outcome: both varieties, and their crosses, are grown everywhere the climate allows.
Why two names under-describe the world
In practice a great deal of tea is neither variety in a pure form. Indian, Sri Lankan and East African planting material is largely hybrid; Taiwanese black-tea breeding has deliberately crossed Burmese large-leaf material with indigenous Taiwanese plants; Japanese black-tea cultivars are assamica-derived selections grown in a small-leaf country. The catalogue therefore carries a `hybrid` variety group and uses it often, and that is a description of ancestry rather than an evasion. It also carries `unknown`, which is the honest label for a great deal of seed-grown material whose composition nobody has established. A reader should read a variety group as a strong prior rather than a determination. It tells you what to expect of leaf size, hardiness and general direction. It does not tell you what a specific field will do, because the plant’s behaviour is a product of variety, cultivar, site, management and season together, and the variety is only the first of those.
What to do with a variety claim
Variety is rarely stated on packaging and is frequently deducible. A Yunnan sheng pu-erh is large-leaf; a Japanese sencha is small-leaf; a Darjeeling is a China-type or a China hybrid; an Assam CTC is large-leaf. Where the claim is stated it is usually being used to explain something specific — why a pu-erh is thick and drying, why a Taiwanese black tea has an aroma no Chinese black tea has — and in that context it is doing real work. Where it is used decoratively, as a synonym for authenticity, it is doing none. Two smaller cautions. The word variety in ordinary English usage frequently means cultivar, or type of tea, or product line, and almost never means botanical variety, so the term on a label is more likely to be loose than precise. And var. assamica is a botanical name, not a geographical one: an assamica plant in Yunnan is not an Indian import, since the variety is native across a range that includes both.