Deciduous and Persistent

So far as I can determine, the terms persistent and deciduous lack clear definitions — certainly not in terms capable of intersubjective application. And it would seem that they are not understood in a consistent manner among botanists. A selection of typical definitions is given below, with significant terms in boldface:

Oxford English Dictionary online
    deciduous:

• Bot. and Zool. Of parts of plants or animals (as leaves, petals, teeth, horns, etc.): Falling off or shed at a particular time, season, or stage of growth. Opposed to persistent or permanent.
• Bot. Of a tree or shrub: That sheds its leaves every year; opposed to evergreen.
• fig. Fleeting, transitory; perishing or disappearing after having served its purpose.

Jepson 1925
    Deciduous

falling when ripe or after the function has been performed; a corolla is deciduous when it falls after anthesis; deciduous trees shed their leaves in autumn

    Persistent

falling away very tardily or not at all

Kearney & Peebles 1951
    Deciduous

Falling, as leaves in autumn, contrasted with evergreen; also when applied to sepals and other organs, contrasted with persistent.

    Persistent

Remaining long attached, not caducous.

Hitchcock et al. 1969
    deciduous

Falling after completion of the normal function. A deciduous tree is one which normally loses its leaves at the approach of winter or the dormant season. (Compare caducous, persistent.)

    persistent

Remaining attached after the normal function has been completed.

Correll & Johnston 1970:(definition by specific example):
    Deciduous.

Falling off, as petals fall after flowering, or leaves of non evergreen trees in autumn or said of plants whose leaves fall, as in "deciduous tree."

    Persistent.

Remaining attached, as a calyx on the fruit.

Martin & Hutchins 1980
    Deciduous

Falling off

    Persistent

Remaining attached longer than typical

Diggs et al. 1999:
    Deciduous

Falling away; not persistent over a long period of time.

    Persistent

Remaining attached; not falling off; contrasting with deciduous.

Harris & Harris 2001:
    Deciduous.

Falling off, as leaves from a tree; not evergreen; not persistent.

    Persistent.

Remaining attached after similar parts are normally dropped, after the function has been completed.

Online Glossary of the Missouri Botanical Garden:
    deciduous

falling seasonally, e.g. of the leaves or bark of some trees, cf. evergreen, also cladoptosis, or, of the calyx or corolla, abscising and falling completely from the flower ....

    persistent

remaining attached to the plant, albeit withered, beyond the expected time of falling, e.g. of sepals not falling after flowering....

Comment

All agree that deciduous involves a process of 'falling,' or somewhat more vaguely, 'being shed.' 'Abscising' adds a structural factor, which would seem to be more essential for botanical description, but not sufficient. Our Berberis species are always described as evergreen (as opposed to Berberis proper: 'deciduous or evergreen'). Yet the leaflets fall at clearly defined abscission zones. (Cf. Disarticulation of Berberis Leaves.) The crucial distinction here being that there is no time when all leaflets fall at once.

Most relate this loss to a vague temporal limit: season, time of year, stage of growth, loss of other organs, 'flowering.' Although timing is, of course, relevant, there is a danger of circularity here if each organ is assigned an ad hoc season determined by the time at which it is shed. After all, virtually every part of a plant is eventually lost. Although popular usage of deciduous when applied to seasonal leaf fall for classes of woody plants cannot be denied, its extension to taxonomic description would seem to require much greater precision than one encounters in normal practice.

Another critical notion is completion of a 'purpose' or 'function' as determining the time at which a plant organ might be expected to fall. But this criterion seems also more difficult to apply than one might anticipate. The primary function of bud scales is usually seen as protecting the bud; e.g., "Most buds are protected by bud scales which fall off as bud tissue begins to grow." If these persist beyond the emergence of a fully developed inflorescence, leaf or stem shoot — as is the case with Berberis/Mahonia bud scales — then unqualified 'deciduous' would not seem an appropriate description.

The original role of the persistent—deciduous distinction in Berberis taxonomy was to uniquely identify B. nervosa (later extended to Asian species) with large and long-persisting bud scales – although size alone would have sufficed. But 'not long-persistent' rather than 'deciduous' is the binary opposite of 'long persistent.' There would seem to be a gradient scale of persistence in the literature:

permanent — long persisting — persisting — generally deciduous — deciduous — caducous

No term is the 'opposite' of the other.