34 MESOZOIC FLORAS OF UNITED STATES. 



afea and the stumps all .stand erect upon it, and many of them are not 

 deeply buried in the earth, but show the natural enlargement toward 

 the roots. A careful examination of these stumps convinced me that 

 thev were not only in place but stood precisely where they grew. Mr. 

 Brown, who discovered this place, is of the same opinion. However 

 difficult it may be to figure to oneself conditions that would preserve 

 trees in an erect position in a sedimentary bed, the fact of their pres- 

 ence in this position seems to Ije conclusive. The most probable theoiy 

 seems to be that, as a matter of fact, the trunks of the trees were not 

 presei'\('d, but only the roots and short stumps, and whatever the con- 

 ditions may have been that were sufficient to preserve large trimks in 

 a horizontal position, the same conditions would surely preserve such 

 short stumps and roots. 



This seems the proper place to mention another phenomenon which 

 at first was very enigmatic and which can not yet be said to l)e adequately 

 explained. I found on several occasions some peculiar short chalce- 

 donized stems contracted at both ends, many of them broken ti'ans- 

 versely and showing a concentric structure; others split longitudinally. 

 I observed that these varied in length relatively to their diameter until 

 some of them became merely elliptical objects resembling fruits. It 

 was not until the erect stumps, above described, were discovered that 

 the mystery was partially cleared up. Among the chips and blocks 

 that surround these stumps there occur a large number of these ellip- 

 tical fruit-like objects, usually striate on the surface and somewhat 

 flattened, so that the cross section is elliptical. This is the typical 

 form and much resembles a butternut that has lost its exocarp, but 

 a yovy little search reveals the fact that there are great variations from 

 this norm, especially in the matter of lengthening the axis. Then it 

 is soon seen by specimens that can be picked up that the rounded ends 

 represent constrictions between two of the objects, and that they are 

 arranged piimarily along a general axis in a necklace-shaped series. 

 The next and most important fact that comes out is that these rows 

 of nut-like objects adhere to the true fossil wood in the interior of the 

 tnmk and are often actually found in place in the stumps as an integral 

 part of their structure. This, of course, reveals their true character 

 as simply accunuilations or secretions of certain suljstances within the 

 trunks, and everything points to the probability that they consisted 



