1859.] REVIEW. 79 



from all the parts of a plants wlietlier the species and genus are 

 to be determined on the artificial or the natural method. 



He next shoAVS that the mode of arranging Orders by the 

 number and disposition or insertion of the stamens is not very 

 much difierent in principle from that adopted by Jussieu^ etc., 

 viz. the number, disposition, or insertion of the external parts of 

 the flower ; — the relations of the calyx and corolla to the ovary ; 

 — the normal divisions of the external envelopes, etc. And he 

 maintains that by strictly following these principles the natural 

 orders are violated. " Si quis illius characteris regulam severe 

 sequetur, naturales affinitates forsan ssepius, quam si hos stricte 

 sequetur [the Linnsean], violabuntur.^' It is further remark- 

 able that the celebrated author of the Natural Orders does not 

 call his invention a natural system. The title of his eminent 

 work is, " Genera Plantarum secundum Ordines Naturales dis- 

 posita, juxta Methodum in horto Parisiensi exaratum." He is in- 

 debted to his followers for the reputation of a natural system- 

 maker. It is universally known that the elder Jussieu arranged 

 his Orders on characters which he believed to be the most con- 

 stant, viz, the Acotyledons, Monocotyledons, and Dicotyledons 

 on the structure of the embryo ; he derived the sections hypo- 

 gyne<2, perigynece, and epigynece, from the insertion of the sta- 

 mens ; apetala, monopetalcB, and polypetalte, from the absence or 

 from the nature of the corolla. This disposition has been fol- 

 lowed by most botanists. Some other characters have been de- 

 tected [inventi), our author says, on which much stress is laid, 

 or to which much importance is attached, and to these we are 

 indebted for the discovery of the natural division of all phseno- 

 gamous plants into angiospermous and gymnospermous ; and to 

 which we owe the new class Dictyogens, the Rhizanths, etc. etc. 

 Subsequently the author insists on the imperfections of this 

 system by its results, viz. that some families may be selected 

 that are sufficiently characterized by the vegetative organs, 

 which are the most liable to mutation ; while a number of fami- 

 lies may be selected in which the law of insertion even is not 

 constant, the Saxifrages for example. In the genus Saxifrage 

 there are examples in which the ovary and calyx are adherent 

 or epigynous, or only partly so, perigynous. Also that the 

 structure of the embryo is not uniform in several families, Nym- 

 pheeacecB, Convolvulacem, Lentibidacea, etc., where this organ is 

 undivided, as in Monocotyledons. 



