others living in a great variety of stations and countries—I have seen no real tendency of H. linariifolium to pass into H. humifusum, still less into H. perforatum.
On the other hand, a very good example of really intermediate forms, erroneously (in my opinion) considered as constituting a distinct species, is afforded by the Daisy. In my early botanical days I was familiar with the two extreme forms—the large-flowered, long-leaved Bellis sylvestris of the south of Prance, of which I dried rather largely, selecting (as is usual with collectors), the most characteristic specimens, and our common, much smaller-flowered, and broader-leaved B. perennis, which I never particularly examined, and which is reckoned too common a plant to be frequently preserved in herbaria beyond a single specimen. The difference between the two was striking; and I adopted, without hesitation or consideration, their established specific distinctness. I subsequently received from Prof. Gussone his B. intermedia, which I laid in, on his authority, as a distinct species, the single specimen being quite insufficient to enable me to form any independent opinion on the subject. But when, in the autumn of 1846, I saw the neighbourhood of Constantinople abounding in daisies of various sizes, usually fully as large as the Montpelier ones, but sometimes much more like our northern ones, and equally variable in the form of their leaves, I felt much puzzled as to which species I should refer them to. In the following spring, in my Sicilian herborisations in Gussone's own country, I paid particular attention to these plants. The three supposed species there appeared to me to pass most gradually the one into the other, the intermediates being more abundant than either of the extremes; and since that time, in other parts of Europe, I have observed that where either of the extremes grows alone, its distinctive characters are not nearly so constant as they are supposed to be. I have thus been irresistibly led to the conviction that Bellis intermedia and sylvestris are mere varieties of B. perennis.
In the above instances, the evidences of specific diversity in the two first, and of identity in the third, are to my mind conclusive; and, as further examples of cases where a conviction of specific identity has, as it were, been forced upon me in opposition either to the views I had at first entertained, or to those of a large number of modern botanists, I would refer to Fumaria officinalis, Cerastium vulgatum, Rubus fruticosus, &c., which have all been the subject of long-contined observation, and endeavours to maintain as distinct species forms which I have, in my Handbook, reunited under the above names. There are, however, a number of cases where the evidences, as hitherto collected, are so insufficient or so conflicting, as to render any satisfactory decision hopeless, until carefully conducted experiments and observations shall have made us better acquainted with the hereditary permanence of certain apparently positive, but minute and unimportant characters.
It may be observed, in the first place, that there are frequently two nearly allied forms, of nearly the same geographical range, which are found more or less in company with each other, retaining over the
