76 Elkins and Wieland — On Devonian Wood. 



In defining new species it is sometimes desirable to state cer- 

 tain facts of negation as succinctly as may be. Accordingly in 

 the present case it is well to repeat that the exact nature of the 

 growth ring especially in forms so ancient as those in question 

 is not fully understood. And since the growth rings described 

 above occur at irregular intervals it is possible that a longer 

 period of growth may have ensued at times between two suc- 

 cessive appearances of the stimulus necessary to cause the 

 formation of growth rings. It is also possible that the speci- 

 men studied by Penhallow may have been taken from an area 

 between two such successive rings of unusual width. This is 

 merely a conjecture. The far greater chance is that both 

 specimens are average trunk wood, and do differ by presence 

 and absence of the growth rings. 



Furthermore Penhallow describes the ends of the ray cells as 

 being conspicuously narrowed. It is true that the ends of the 

 ray cells in the Indiana specimen are often narrower than the 

 rest of the cell, but it is not a constant feature and is never 

 conspicuously true. The reader may recall that in the descrip- 

 tion of the ray cells it was stated that the horizontal walls of 

 the outer rows of cells were frequently irregular and appeared 

 to narrow at the ends when only returning to the normal cell 

 diameter, a condition which might be misleading. There are 

 other minor differences between C. Newberryi and the Indiana 

 form in regard to the number of pits in a group, the height of a 

 ray, etc., which could be accounted for in a species with very 

 wide limits of variation, as seen from the preceding table. 

 But one may only couclude that in the absence of evidence 

 that the differences apparently present are simply variations 

 within the species, the logical course is to erect a new species 

 based on good material. 



In dealing with the present fine fossil tree from Indiana it 

 has been preferable so far to use the old names applied to wood 

 of Cordaitean affinity ; but in naming the new species quite 

 clearly indicated to be present it appears necessary to include 

 it together with the Cordaites Newberryi in the new genus, 

 Callixylon, recently established by Zalessky* with apparent 

 correctness for Cordaitalean woods with grouped bordered pits. 

 A new Russian form he describes, Callixylon Trifilievi, from 

 the Upper Devonian of Bolchaia-Karakouba, district of 

 Marioupol in the province of Ekaterinoslaw, has this character. 

 Only the preliminary description of this Russian species with 

 three photomicrographs has so far come to hand. But Zalessky 

 shows that while the Araucarian type of pitting is present, and 

 the grouping is distinct as in our two American forms, the 



* Etude sur l'Anatomie du Dado.rylon Tchihatchefii Goeppert sp., par M. D. 

 Zalesskv. Memoires du Comite* Ge"ologique de Russie, Nouvelle Serie, Liv. 

 68, St. Petersbourg, 1911, pp. 29 with 4 plates. 



