deduce the order of the genera theoretically, in order to group them into families, he imitated the manner in which the genera themselves were formed. Botanists, struck with the complete and constant resemblance of certain individuals, had collected them into species; then, according to an equally constant resemblance, but much less complete, had collected the species into genera. The characteristics, which may vary in the same species, will depend on causes not innate in the plant, such as its height, the hardness or softness of its wood, certain modifications of shape and colour, &c., which change with the soil, the climate, and other purely accidental influences. The specific characteristics, on the contrary (those which ought to be presented by every individual, that is connected with others in form-. ing a certain species, whatever may be the circumstances in which it is placed), will be inherent in the very nature of the plant. Amongst these characteristics there are some more important than others, less subject to vary in the different individuals; these, being always found in a certain number of species, impress upon them a resemblance sufficiently striking to allow us to constitute a genus, These will therefore have more value on account of their generality than the specific, and the specific than the individual. how can we appreciate these different values? Nature herself has indicated to the observer the species and several of the genera by the points of resemblance with which she marks certain vegetables; beyond these genera this conducting thread was wanting, since all botanists, agreeing in almost everything up to this point, differed after they reached it, and followed each a separate route. There are, however, several large groups of vegetables connected with one another by characteristics of resemblance so evident, that they cannot escape the notice of the most casual observer, much less of a botanist. Besides these points of resemblance, common to every species of one of these groups, there are some which are only common to a certain number among them; so that it may be subdivided into a large number of secondary groups. These had been recognized as genera by But botanists. There were, therefore, already a few collections of genera evidently more similar to one another than they were to those of any other group, or, in other terms, some families undeniably natural. Jussieu thought that this was the key of the natural method, since, by comparing the characteristics of one of these families with those of the genera which compose it, he would obtain the relation of one to the other; since, by comparing several of them with one another he would see what characteristics, common to all the plants of the family, varied in such a one and such another; since he would thus arrive at the value of each characteristic, and this value, once determined by means of these groups so clearly arranged by Nature herself, could in its turn be applied to the determination of those on which she has not so clearly imprinted this family likeness, and which were the unknown quantities in the great problem. He chose, therefore, seven families universally admitted; those, which are known under the names of Gramineæ (Graminées), Liliaceæ (Liliacées), Labiata (Labiées), Compositæ (Composées), Umbelliferæ (Umbellifères), Crucifera (Crucifères), and Leguminosa (Légumineuses). He discovered that the structure of the embryo is identical in all the plants of one of these families; that it is Monocotyledonous in the Gramineæ and in the Liliaceæ, Dicotyledonous in the five others; that the structure of the seed is also identical; the Monocotyledonous embryo is placed in the axis of a fleshy perisperm in the Liliaceæ, on the side of a farinaceous perisperm in the Gramineæ; the Dicotyledonous embryo, at the summit of a hard and horny perisperm in the Umbelliferæ, without a perisperm in the three others; that the stamens, which may vary in their number in the same family, the Gramineæ, for instance, do not generally vary in the method of their insertion, Hypogynous in the Gramineæ and in the Cruciferæ; on the corolla in the Labiata and the Composite; on an epigynous disk in the Umbelliferæ. He thus obtained the value of certain characteristics which would not vary in the same natural family. But, less in importance than these, there were others more variable, which he tried to appreciate in the same way, either by the study of other families formed by Nature herself, or in those which he formed by applying these first rules and several others, also founded on his observations. We cannot here enter into the details of this long and arduous undertaking, from which resulted a hundred families containing all the plants known at that time."-p. 580. K. Reply to the Editorial Observations on the Robertsonian Saxifrages, at page 451, &c. By CHARLES CARDALE BABINGTON, Esq., M.A., F.L.S., &c. In reply to the remarks of "C." in the 'Phytologist' (Phytol. iii. 451, 452), I wish to state that there has never been any desire on my part to avoid the acknowledgment of a "mistake" of mine, which indeed was not a mistake at the time of its publication. As the new VOL. III. 3 Q facts observed by Mr. Andrews had been published in the report of the meeting of the London Botanical Society, of April 4, 1846 (Phytol. ii. 537; Gard. Chron. 1846, p. 254), at which he proved, by the exhibition of a series of specimens, that "the fact of the Hibernian forms including those of the Pyrenees is placed beyond all doubt or cavil," I never suspected that I was bound to write a paper upon the subject, nor that any charge could be brought against me, even by your correspondent "C.", for not going out of my way to publish those facts which Mr. Andrews was so well qualified to publish himself, and indeed had published. I supposed that all the interest of the subject was at an end after Mr. Andrews' discovery of the Pyrenean forms of the Robertsonian Saxifrages in Ireland, as that interest consisted in their supposed absence. In answer to a letter from me, I am informed by Dr. Harvey that Mr. Andrews thinks that I prevented the publication of part of his remarks in the Reports of the British Association, and also of a paper sent to the 'Annals of Natural History' by him. With the publication of the former I had nothing to do, as the local officers of the Association are not consulted upon such matters, and have no hand even in transmitting the papers, communicated to the sections, to head quarters. The report was probably drawn up by one of the Secretaries of the section, to whom abstracts of papers ought to be given by their authors. Had that been done in this case, the report would have been such as Mr. Andrews wished. Concerning the paper sent to be inserted in the Annals I know nothing, as it is not usual for the Editors of that Journal to submit papers, reflecting upon, or contradicting the statements of, one of their own body, to the person upon whom the reflections are made. I therefore did not see it, and know nothing about it. I feel certain that a short statement would have been printed. After what had appeared in the 'Phytologist' (Phytol. ii. 537), I thought that I had done all that was requisite by adding, in the 2nd edition of my Manual (p. 126), which was published in 1847, to the account of Saxifraga umbrosa the words "All the forms are found in the west of Ireland," thus including amongst those found in Ireland the a. crenata, which was stated in the 1st edition to be the Pyrenean plant. What I stated "so positively" in the Annals was, that having had occasion to re-examine the Irish Saxifrages, I had been "greatly struck by the uniform difference which exists between each of them and the corresponding plant of the Pyrenees." In this, all that requires alteration even now, is the substitution of the word "usual " for "uniform ;" for I still think that the usual Irish forms differ from the usual Pyrenean forms of the plants. I am truly sorry that any omission on my part should have appeared like a neglect of Mr. Andrews' observations, and am not surprised at his feeling sore when he supposed that it was intentional. The existence of the Pyrenean forms in Ireland has now, at any rate, been made sufficiently public by appearing in two of the Botanical Journals, in addition to the report referred to above. St. John's College, Cambridge, CHARLES C. BABINGTON. A few Remarks on the "Proof" of C. C. Babington's "Error” respecting the Specific Distinctness of Saxifraga Geum, elegans, hirsuta, &c., &c. By JAMES BACKHOUSE, JUN., Esq. WHILST the information given by "C." in the last No. of the 'Phytologist,' upon the Saxifrages of the umbrosa group in Ireland, cannot be regarded as otherwise than interesting and important, inasmuch as specimens have been gathered, seeds collected, and plants cultivated from both Ireland and the Pyrenees in abundance; yet the decision which "C." seems to have arrived at, that there is (or has been) an amount of evidence given sufficient to overthrow the specific distinction between umbrosa, elegans, hirsuta, and Geum which many persons believe to exist; and even to require from a careful and accurate observer, like C. C. Babington, as a "scientific, if not a moral, obligation," that he should retract his statements and declare his error, seems in my opinion entirely devoid of sufficient foundation and authority. I cannot say that the probabilites do not appear to be on the side of those who think these Saxifrages varying forms of one species; but before any certain conclusion is arrived at, we must be in possession of far more evidence than has yet been brought forward in the public journals. The information respecting them actually ascertained, seems to be, that every shade of difference between the extreme forms (which may be considered as represented in S. umbrosa, var. serratifolia, and S. Geum, setting aside the "new one?") is found in a wild state, not only in Ireland, but also in the Pyrenees; and that seeds have been collected, the plants raised from which have likewise passed completely one into the other; also that under cultivation the same variable characters are exhibited. These are important points; but now the question presents itself, WHENCE ARISES THIS INFINITE VARIETY? It may be that there is one species only, which, when far removed from others and every possibility of hybridization, has a natural tendency “ to sport ” in the seedlings, giving rise to an endless variety. On the other hand it may be that there are three or four species, or more, forming a group, which are readily affected by hybridization, the seedlings from any one of which may, under such influence, show every imaginable intermediate state.* Or it may be that there are many species, the individuals of which have a tendency, without hybridization, to vary and "sport" from seed; the seedling from one species so closely approximating the seedling from another species, as scarcely to be distinguishable. It is only necessary to refer those who have examined extreme forms of Polystichum lobatum, var. (?) lonchitidioides, to its close resemblance to P. Lonchitis; and yet it is known that the former under cultivation assumes the common form of P. lobatum, while on the other hand P. Lonchitis retains its character, or quickly regains it, where the original plant has taken the form of P. lobatum "lonchitidioides," by having grown in a shaded place. Till these intricate questions are settled, who can be justified in saying that such are species or such are only varieties, or much more in saying that there is a "moral," or even "scientific, obligation" for a person holding one view to declare his error? Because seeds procured from wild plants in Ireland or the Pyrenees produce infinite variety, we have no proof against the existence of several species: and because intermediate forms are found abundantly in the Pyrenees as well as in Ireland, it may be answered by way of argument, that like causes under like circumstances produce like effects. Until extreme forms have been cultivated, singly and far apart from others of the group, for a season or two, and seedlings have been raised from these, which not only vary, but actually produce the opposite extreme, it appears to me that no one can say with certainty that there are not several species which have a tendency "to sport," * As in the case of Pyrus Aria and P. Aucuparia, where by hybridization every intermediate form has been produced. |