222 FLORA OF LOWER COAL MEASURES OF MISSOURI. 



while generally the central trace in the roundish area is more or less clearly- 

 shown. This is also true where the outermost cortical tissue still adheres 

 to the matrix, even in the area represented by the shaded portion in the 

 original figure. 



Another of the types described by Lesquereux is a small slab, No. 

 5502 of the Lacoe collection, representing, like the other, a mold or 

 impression of the stem. In a portion of this fragment, too, we have 

 bolsters and inside areas like those in the originally figured type, No. 

 5501. But here we have also a variety of distortions, due to pressure, 

 in which the central areas or compressed bosses often appear more than 

 twice as wide as long, while in some cases they are partially covered on 

 all sides by the infolded lateral areas about the flattened bosses. It may 

 be noted in passing that the central areas in this fragment, photog'raphed 

 in PL LXVI, Fig. 1, are much broader in proportion to their altitude when 

 the bolsters themselves are correspondingly dilated. In addition to these 

 features, this specimen shows not only the central traces, but also, in a few 

 cases, the obscure vertical furrows, which in several instances seem to unite 

 below the upper margin of the central areas in a loop or long horseshoe, 

 between the sides of which are the central traces. This character, as well 

 as certain other more obscure details, will be considered in the description 

 of the surface of the lately collected stems. 



Another incomplete slab about 25 cm. wide, showing the mold or 

 external impression of the stem, is partially illustrated in PI. LXVI, Fig. 2. 

 In this specimen, chosen because it represents the more elongated bolsters 

 witli the central convex areas slig'htly displaced, we find in many of the 

 latter the two vertical grooves, about 2 mm. apart, passing' across the 

 compressed boss and forming, as in the specimen just described, a loop or 

 elongated horseshoe, within which th6 vascular trace is seen in all cases to 

 lie. Occasionally a second trace is observed at the apparently open end of 

 the horseshoe. The same interior characters are seen in Mus. Reg. 6030, 

 another impression of a fragment with short, squarish bolsters, illustrated in 

 PI. LVIII, Fig. 2. 



The specimens described above are all impressions or molds of stems, 

 in some of which the ejaidermis may have been wanting. We will now 

 proceed to the consideration of several segments of stems on which the 

 cortex is still preserved. The first of these, Mus. Reg. 6029, is a flattened 



