IM ;;.'). If. i). CLASSIFICATION OF FOSSIL I'LANTS. 381 



like those of Sji/nnoh /lii/imii, FJiitnhs, and (jIIkt guiuTci', caii- 

 iiut be re^j^arded m ijfso as an <)l)jection ai^ainst its usj*. It 

 is, on the contrarv, the casp that th«* actual occiirreuce in the 

 same specimens of Ixtth these types iui|dird the use of the 

 same name for Ixith of them. W'lien we take into consideraticjii 

 the further fact that the two ty[)es are by no means distinct 

 — the dorsiventrality of the yV/:ri7c.s-shoots ncjt being due to 

 the arrangement of the leaves on the shoots, which is onl\- 

 pseudo-distichous — there won hi apjjcar to l)e good reasons for 

 using only one name. 



The most important objection against the use of Elatocladus 

 as I have proposed it, and one of which I am well aware, is 

 that it will have to include a large number of forms which 

 really may be very diiFerent. But this is due to the nature 

 of the material: when the latter does not show any sufficient 

 gaps in the series of variations to permit the separation of 

 dilt'erent form-genera, the best thing seems to l)e to give way 

 before this fact. The genus is in a high degree a provisio- 

 nal one and should be used only for forms which it is im- 

 possil)le to place in a better defined form-genus. Some such 

 form-genera already exist, and many more will no doubt have 

 t(j be distinguished in the future. 



In bringing these remarks to a close it may be once more 

 emphasized that it is not intended to advocate that the deter- 

 i minations discussed here are necessarily right. All determi- 

 nations of impressions of fossil plants must naturally be 

 ])rovisional and errors ai*e inevitable, whichever method id' 

 classification is adopted. Of this, however, there can be no 

 doubt, that the errors bring much less evil consecjuenees in 

 the way (d" confusion and uncertainty if, within reasonable 

 limits, the species are kept small and as strictly defined as 

 possible. 



