whole of that range certain distinctive characters, of no great importance in their respective genera, yet apparently constant in that particular case. Such are, for instance, Viola odorata and hirta, Lychnis vespertina and diurna, Ulex Europæus and nanus, Sonchus oleraceus and asper, Senecio jacobæa and erucæfolius, Orchis maculata and latifolia, Juncus articulatus and obtusiflorus, and a number of others. In some of these cases, the balance of evidence has appeared to me to be in favour of their specific distinctness, in others of their identity, and I have so recorded them in my Hand-book, but often with great hesitation; and it is not improbable that further observation and experiment may induce a change of opinion in regard to some of them.
Again, there are sometimes two, three, or more forms, having every appearance of really distinct species, all common over an extensive area, or spreading into distant regions, and everywhere retaining their characters; and yet we are occasionally startled by the appearance of intermediate forms of various degrees, suggesting in some minds the specific identity of the whole series, in others a progressive development from one species to another; and in others, again, natural hybrids; whilst in some instances the observer may have been deceived by accidentally abnormal specimens, carefully preserved and occupying a conspicuous part in the herbarium, without any record of the attending circumstances which might have accounted for their production, but which forms in nature are very rare, and of mere temporary existence. Such occur, for instance, among some of the common species of Rumex, Mentha, &c. It is also frequently a matter of great nicety to determine what constitutes an intermediate form; for two plants, to be really intermediate, should not be so in one character only,, but in general habit and aspect, in a combination of all the characters which separate the two species it stands between. The species of Carduus (including Cirsium) for instance, have been artificially divided into species, with their leaves decurrent or not. When, therefore, a specimen of one which has usually sessile leaves is met with having them slightly decurrent, it has been, on that account alone, set down as intermediate between that and some other species to which it shows no approach in any other point; and thus figures in books as a hybrid, or a distinct species, according to the tendency of its describer.
One source of deception as to the real permanency of an abnormal form, even when observed without variation in a wild state in the greatest abundance, arises from the facility with which certain perennials, or shrubs, multiply by runners, suckers, bulbs, or other modes of division, especially in cool, moist, and comparatively sunless climates like our own. Individual peculiarities are thus propagated naturally in a wild state, as we do artificially in gardens, spreading over the country in such numbers, as to be mistaken by the cursory observer for races, if not for species. Seedling brambles, mints, creeping-rooted weeds, &c., are rare in our climate; the bulbiferous Alliums, the viviparous grasses, many introduced plants, such as the Periwinkles, Hypericum grandiflorum, &c., seldom produce any seed. Carduus arvensis
