176 W. T. GORDON 
P]. III. fig. 27 represents a cross-section of another of these petiole fragments which 
is exactly similar to that of fig. 28. The one end is open and the other closed by a 
pinna-trace-bar in exactly the same condition as in the latter figure. Ata higher level 
in the specimen the pinna-trace-bar became detached from the petiole-trace and showed 
some signs of dividing into two (PI. II. fig. 26, b). A twisting of the whole petiole- 
trace at this level gives a much longer appearance than is shown in Pl. III. fig. 27. In 
the next figure (fig. 25) an exactly similar section is represented. The plane of section 
in this case is horizontal, except at the left-hand side. In this specimen there is also a 
reduced pinna-trace-bar given off from the petiole-trace, and, as in the case of fig. 26, the 
bar shows signs of dividing in the middle to form two traces. 
Passing to the trace represented in Pl. II. fig. 24, the first thing to be noticed is 
that it has suffered no distortion and that it is essentially similar to that of fig. 25. In 
general form it greatly resembles Pl. III. fig. 29, but it is larger and the pinna-trace-bar 
b at the top of the figure is much better developed than in the former specimen. The 
petiole from which the section was prepared is about 6 inches long, and a little above 
the level of this section two pinna-traces were noted. They were fairly well developed, 
but did not penetrate the cortex to enter into distinct pinnee. The outer cortex Ce ¢.) 
in the figure is also worthy of notice, since it contains no sclerotic layer. 
A very similar example is shown in Pl. II. fig. 23. The trace is not quite so long 
as the last, but it is much stouter, and the bar b is rather better developed. Yet this bar 
also never gives rise to traces which enter into distinct pinne. In fig. 23 it is practically 
divided into two, but both parts die out higher up. The outer cortex in this specimen 
had a distinct sclerotic band towards the periphery. Up to this point all the traces 
examined would have been referred to Dineuron had they occurred separately, and 
indeed would still be referred to that genus unless the next three stages had been 
discovered. The first of these transition forms is shown in Pl. II. fig. 22. It is 
essentially like that of fig. 28, but a flattening is making its appearance in both sides 
and the pinna-trace-bar is rather well developed. A similar flattening may be noted in 
Pl. IL. fig. 26, but this is largely due to the obliquity of the plane of section, which causes 
an apparent elongation of the trace. PI. II. fig. 21 exhibits another of these transition 
stages. Here the trace gives some indication of a waist. The pinna-trace-bar, also, is 
much more robust than in the last figure. This trace is exactly intermediate, both 
in size and shape, between the Dinewron-like example of fig. 24 and the normal 
Metaclepsydropsis trace of figs. 17 and 18. 
In the specimen shown in fig. 20 we have the last of the transition types. A 
distinct waist is shown in this example, and therefore the trace, in transverse section, 
has the appearance of an hour-glass. An early stage of pinna-trace departure is also 
demonstrated by the specimen, and the bar is quite robust and well developed. Probably 
even this specimen would only be referred to Metaclepsydropsis duplex with some 
misgivings. Fig. 19, however, supplies the last link in the chain, and, while the trace 
shown in it would at once be accepted as typical of Metaclepsydropsis duplex, there 
