Secr. XX. 7. NATURAL CLASSES. 577

ricties of its figure, as cylindrical, angular, awled, capillary; and to thefe were again added the divifions of the ftigma, as convolute, re- volute, fix-parted, many-parted. And to thefe were again added the various forms of the ftigma, as globular, egsed, end-nicked, cruciform, feathery,&c. which are enumerated in the Philofophia Botanica; there is great reafon to believe, that chara@teriftic marks of all the orders of plants might be deduced and named from fome of thofe circumftances feparately or conjointly; which might diftin- guifh them from each other with-oreater eafe and certainty, and by raarks lefs variable by foil or climate, than by the number alone; and by rendering them more natural add to the beauty and utility of the Linnæan fyftem.

Concl u/ion.

Neverthelefs I am well aware of the great general inconvenience of altering fo extenfive a fyftem once eftablifhed, and am forry to fee. fome idle efforts to add the clafles already deduced from fituation or proportion to thofe, which are fimply numerical; and thus rather to deteriorate than to improve the prefent fyftem of the great mafter.

T profefs myfelf incapable to execute the plan, which I have here fuggefted, as it would require a moft exact knowledge of the detail of botany, as well as of the outline; would require many years of un- remitted application, with every opportunity of vifiting botanic gar- dens, or examining dry colle@ions, and infpe@ting prints and draw- ings of vesetables; and would demand a genius, which few poflefs, capable of reducing the complex and intricate to the fimple and ex- plicit.

But 1f the fyftem of the great Linnæus can ever be intrinfically improved, I am perfuaded, that the plan here propofed of ufing the fituations, proportions, or forms, with or without the numbers of

à À the

ET

à 1 Ÿ ? Hi