40 IOWA ACADEMY OF SCIENCES. 



he finds, among the four new species studied, two forms which are clearly dicoty- 

 ledonous, and two other distinctly coniferous m relationship. The species are: 



Coniferous. Dicotyledonous. 



Cupressinox)jlon arkansanum, Lcmrinoxijlon hranneri, 



Cupressinoxi/Ion colli. Laurinoxylon lesquereuxiana. 



There was also a single additional specimen whose affinities appeared to be 

 dicotyledonous and to belong to Laurinoxylon; the condition ol the material would 

 not admit of a clo.ser determination. The specimens found indicate comparatively 

 few species, but these few must have existed in great numbers. One of the most 

 valuable and pertinent facts in this connection is the finding of the dicotyledonous 

 Laurinoxylon hranneri in the lignite bed of Bolivar Creek, as lignite, deeply 

 buried in Eocene clays in massive form. 



Thus far sufficient distributional facts to give a taxonomic value to the fossil 

 woods have not been discovered. Until extensive collections throughout the whole 

 region of the southern Tertiary have been made it will not be possible to use these 

 forms for purposes of diflferentiation or of correlation. It is believed, however, that 

 since in the Tertiary sands of Arkansas, Louisana, Texas and Mississippi the same 

 relations of silicified woods to lignites have been observed, it may be possible to 

 co-ordinate the divisions recognized in those States by geologists and devise a 

 system of nomenclature that will explain the relationships of the various beds to 

 each other, though it cannot be done at present. 



During the progress of the study of the region by the writer it became more and 

 more clear that the silicified wood had some intimate relation to the pockets or 

 beds of lignite which are scattered throughout the ridge. It was early noticed 

 that no lignite occurs in the sands or gravels above the clays, and that no detached 

 masses of silicified wood occur entirely in the clays. As the investigation proceeded 

 it became a favorite hypothesis that the silicified wood was transformed lignite, 

 and that careful microscopic study would probably prove the hypothesis to be cor- 

 rect. Professor Knowlton's investigations appear to verify the hypothesis. 



The opinion that the silicified wood was, in some way, to be connected with the 

 lignites of the bed underlying the sands was suggested by Hilgard* many years 

 ago. Speaking of the occurence of fossils in the Orange sands he says: "... . 

 The closest scrutiny I have bestowed on hundreds of extensive exposures, has 

 failed to detect any fossil apparently peculiar to the formation as such. This 

 might seem paradoxical enough to anyone acquainted with the frequent occurrence 

 of silicified wood in these strata, but it soon becomes quite obvious to an attentive 

 observer that the regions of the frequent occurrence of this fossil in the Orange 

 sand are coextensive with those in which fossil wood, either silicified— when 

 imbedded in siliceous sands— or lignitized, occurs in the underlying lignitiferous 

 Cretaceous or Tertiary strata. It is not unusual to find trunks of silicified wood 

 imbedded partly in the unchanged lignitic strata, partly in the Orange sand; the 

 portion contained in the latter being nearly or wholly deprived of carbon, while 

 the part imbedded in the lignitic material is, if at all silicified, of an ebony tint 

 and often contains pyrites." Again, "I am convinced that the greater part, if not 

 all of this fossil wood, is derived from the underlying strata and will be represented 

 in their flora." 



There can be little question, therefore, that the process of silifieation has occur- 

 red, in some cases at least, since these masses were torn from the underlying beds 



American Journal of Science, II, Vol. XLI. p. 313. 18l>6. 



