ject at different levels from the bluff and making angles with the present slope.
It is rather remarkable that only one standing stump was seen with a limb in position. This is probably explained by the fact that the living trees were generally covered by the volcanic material to a less height than that of their lowest limb, and consequently the upper portions of the trees were not preserved, but suffered aerial decomposition. In general, the silicified tree would crumble down as rapidly as the rock material surrounding it would wear away, so that only short stumps would now be found, though greater lengths were petrified. The absence of limbs in position is, however, mainly due to the fact first named. In the cases of trees that were petrified after they had fallen, both limbs and roots projecting upward were seen in position.
Specimens of rotten wood far progressed toward complete decomposition were found perfectly preserved in stone. Petrifactions of bark were of frequent occurrence, and the channeling and borings of worms or insects were beautifully preserved in some of the specimens, so that we literally have petrified wormholes.
In some of the finer water-collected debris were found beautifully preserved impressions of leaves, showing two kinds of deciduous trees, of course entirely different from any trees now growing in the region. The impressions of conifer leaves and the petrified part of the same wood were also found.
These fossil tree remains are found over a wide area in the park region. Along Soda Butte Creek they stand up the slope from each bank, but along the Lamar River, below the mouth of this creek, they exist on the left bank only, the imbedding material having been entirely removed from the right bank by erosion. The lowest level at which a petrified tree was seen in position was on the left bank of the Yellowstone, opposite the mouth of Hell-roaring Creek, at an approximate altitude of G,100 feet. The highest was seen opposite the mouth of Soda Butte Creek at an altitude of about 8,180 feet. These trees are twelve or fifteen miles apart, and the original slope of the ground between them is not known, so that they can not be taken to fix the highest and lowest levels of the original forest growths in this area.
At Specimen Ridge, where the closest examination was made, the lowest stump seen in position was at an altitude of about 7,000 feet, and the highest a little over 7,500 feet. There were here between these limiting growths certainly nine successive forests, and of course an equal or greater number of incursions of imbedding materials.
In what has gone before I have not attempted to designate